Embodiment
Below in conjunction with accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Ground describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.Generally exist
The component of the embodiment of the present invention described and illustrated in accompanying drawing can be configured to arrange and design with a variety of herein.Cause
This, the detailed description of the embodiments of the invention to providing in the accompanying drawings is not intended to limit claimed invention below
Scope, but it is merely representative of the selected embodiment of the present invention.Based on embodiments of the invention, those skilled in the art are not doing
The every other embodiment obtained on the premise of going out creative work, belongs to the scope of protection of the invention.
It should be noted that:Similar label and letter represents similar terms in following accompanying drawing, therefore, once a certain Xiang Yi
It is defined, then it further need not be defined and explained in subsequent accompanying drawing in individual accompanying drawing.Meanwhile the present invention's
In description, term " first ", " second " etc. are only used for distinguishing description, and it is not intended that instruction or hint relative importance.
As shown in figure 1, it is the schematic diagram that server 100 provided in an embodiment of the present invention interacts with local terminal 120.
The server 100 is communicatively coupled by network and one or more local terminals 120, to enter row data communication or friendship
Mutually.The server 100 can be the webserver, database server etc..The local terminal 120 can be PC
(personal computer, PC), tablet personal computer, smart mobile phone, personal digital assistant (personal digital
Assistant, PDA) etc..Wherein, client can be installed in local terminal 120, such as browser or some application programs
Deng user can carry out message reference by client.
Fig. 2 shows a kind of structured flowchart for the server 100 that can be applied in the embodiment of the present invention.As shown in Fig. 2 clothes
Business device 100 includes memory 102, storage control 104, and one or more (one is only shown in figure) processors 106, peripheral hardware connect
Mouth 108, radio-frequency module 110 etc..These components are mutually communicated by one or more communication bus/signal wire 116.
Memory 102 can be used for storage software program and module, such as the information-pushing method in the embodiment of the present invention and
Programmed instruction/module corresponding to device, processor 106 are stored in software program and module in memory 102 by operation,
So as to perform various function application and data processing, such as information-pushing method.
Memory 102 may include high speed random access memory, may also include nonvolatile memory, such as one or more magnetic
Property storage device, flash memory or other non-volatile solid state memories.Processor 106 and other possible components are to storage
The access of device 102 can be carried out under the control of storage control 104.
Various input/output devices are coupled to processor 106 and memory 102 by Peripheral Interface 108.In some implementations
In example, Peripheral Interface 108, processor 106 and storage control 104 can be realized in one single chip.In some other reality
In example, they can be realized by independent chip respectively.
Radio-frequency module 110 is used to receiving and sending electromagnetic wave, realizes the mutual conversion of electromagnetic wave and electric signal, so that with
Communication network or other equipment are communicated.
It is appreciated that structure shown in Fig. 2 is only to illustrate, server 100 may also include than shown in Fig. 2 more or more
Few component, or there is the configuration different from shown in Fig. 2.Each component shown in Fig. 2 can use hardware, software or its group
Close and realize.
First embodiment
A kind of information-pushing method is present embodiments provided, for pushing possible information interested to user, is referred to
Fig. 3, this method include:
Step S110:The information that user to be pushed accessed is obtained, as specify information.
User to be pushed is the user to its pushed information, can be not particularly limited as who user.
In all information in database, some information are probably the information that user to be pushed accessed, for example, user uses
News is checked in certain news application program, in all news in database corresponding to the news application program, some news can
Can be that user's history accessed.The information that user accessed is typically user's information interested, so, it can first obtain and treat
The information that push user accessed, specify information is used as using the information.
The access time of some information accessed due to user to be pushed may be from push current time farther out, may not
Agree with very much the current interest-degree of user, or the information that user's history accessed may be more, it is preferred, therefore, that in this implementation
In example, information that user to be pushed accessed in preset time period can be obtained as specify information.For example, when obtaining current
Between user to be pushed accessed in previous time-of-week all information as specify information.Certainly, the dental pulp period is specially
How long, be specially which not as the limitation in the present embodiment period.
Alternatively, it is also possible to obtain the information for the default number of branches that user to be pushed finally accesses as specify information, that is,
Say, obtain the information for the default number of branches that the to be pushed user nearest from current time accessed as specify information.For example, obtain
500 information that user to be pushed finally accesses, as specify information.
In the present embodiment, specifically obtain all information or partial information in the information that user to be pushed accessed,
Information that user to be pushed accessed etc. how is obtained to be not intended as limiting.It is combined to come alternatively, it is also possible to a variety of acquisition modes
The information that user to be pushed accessed is obtained, for example, when the information bar number that user to be pushed accesses in preset time period is less than
During certain amount, information that user to be pushed accessed in preset time period is obtained as specify information;As user to be pushed
The information bar number accessed in preset time period be more than one surely quantity when, can obtain and nearest from current time treat that push is used
The information for the default number of branches that family accessed is as specify information.
Step S120:According to the intensified learning similarity factor between each specify information and each information to be pushed, acquisition is respectively treated
Push coefficient corresponding to pushed information, wherein, the intensified learning phase between any specify information and any information to be pushed
The operation data of the information to be pushed is obtained according to the user for accessing the specify information like coefficient.
Information to be pushed is the information that possible pushed to user, and the information pushed to user is selected from information to be pushed
Take.Information to be pushed can be except specifying letter in all information or database in the database where specify information
Other information beyond breath, information to be pushed are specially which information does not limit in the present embodiment.In addition, information to be pushed
Can be same category information with specify information, if specify information is video, information to be pushed is also video;Specify information is new
Hear, information to be pushed is also news;Specify information is application, and information to be pushed is also application.Certainly, information to be pushed is with specifying
Whether information is that same category information does not limit in the present embodiment.
Different information to be pushed, the interest-degree of user may be different, can be by pushing coefficient quantization in the present embodiment
User pushes to the interest-degree of information to be pushed so as to treat the information of push user by pushing coefficient guidance.
The information accessed due to designating the information as user, may relatively user interest-degree, it is possible to according to
Relation between specify information and information to be pushed, obtain push coefficient corresponding to each bar information to be pushed.
In the present embodiment, can according to the intensified learning similarity factor between each specify information and each information to be pushed,
Obtain push coefficient corresponding to each information to be pushed.
Wherein, the intensified learning similarity factor between any specify information and any information to be pushed was according to accessing
The user of the specify information obtains to the operation data of the information to be pushed, that is to say, that refers to if there is other users to access this
Determine information, then can obtain the specify information to operation data caused by the operation of the information to be pushed according to other users with being somebody's turn to do
Intensified learning similarity factor between information to be pushed.Specifically, various operations pair of the user to information can be previously provided with
The operating parameter answered, for any specify information and any information to be pushed, it can will access the user couple of the specify information
Operation data of the operating parameter corresponding to the operation of the information to be pushed as the user to the information to be pushed.
Specifically, the user pre-set can be to operating parameter corresponding to the various operations of information:In user to referring to
In the accession page for determining information, information to be pushed is pushed to the user, the user does not click on the information to be pushed, the operation pair
The operating parameter answered is the first numerical value;In user in the accession page of specify information, information to be pushed has been pushed to the user,
The user clicks on and have accessed the information to be pushed, but to the information to be pushed without other behaviors, operating parameter corresponding to the operation
For second value, wherein, other behaviors can be commented on, share, thumbing up;In user in the accession page of specify information,
Information to be pushed is pushed to the user, the user is clicked on and have accessed the information to be pushed, and the information to be pushed is generated
Other behaviors, operating parameter corresponding to the operation are third value.In the present embodiment, the first numerical value, second value and
The occurrence of three numerical value is not intended as limiting, and the different interest to the information to be pushed that can be shown according to different operations are true
Fixed, for example, the first numerical value can be -1, second value can be 1, and third value can be 2.
Wherein, any one operation in above-mentioned three kinds of operations can also be divided into multiple operations, set different behaviour respectively
Make parameter, for example, in user in the accession page of specify information, information to be pushed has been pushed to the user, the user clicks on
The information to be pushed is have accessed, and other behaviors are generated to the information to be pushed, other different behaviors set different numerical value
Operating parameter.Furthermore it is possible to according to the issuable operation setting operating parameter of specific information category, for example, for example, letter
Cease for video, in user in the accession page of specify information, information to be pushed has been pushed to the user, user click access
The information to be pushed, user are different to the viewing time of information to be pushed, and the operating parameter of setting is different, and it is possible to be,
Viewing time is longer, and the operating parameter numerical value of setting is bigger.In the present embodiment, the operation that user may be to information does not limit
For the operation in the case of above-mentioned three kinds.
Further, due to that may have different similitudes between different information, user is interested in an information
In the case of, may pair information similar to the information it is also interested, therefore, in the present embodiment, information to be pushed with specify letter
Intensified learning similarity factor between breath, can be according to the similarity system between the operation data and information to information to be pushed
Number is common to be obtained.Wherein it is possible to be, for any specify information and any information to be pushed, accessed this according to all and referred to
The user of information is determined to operation data caused by the operation of the information to be pushed and the information to be specified and the information to be pushed
Between coefficient of similarity, calculate the intensified learning similarity factor between the specify information and the information to be pushed.
Specifically, for any specify information and any information to be pushed, this can be calculated by equation below and is specified
Intensified learning similarity factor between information and the information to be pushed:
Wherein, information j can represent any one information to be pushed, and information i can represent any one specify information,
simI, jFor the coefficient of similarity between information i and information j.V represents all users, feedbackV, jRepresent that user v treats push
Operation data caused by information j operation, wherein, if in user v in information i accession page, non-pushed information j,
feedbackV, jThe 4th numerical value can be entered as, such as 0.viewV, iRepresent whether user v accessed information i, user v is as access
Specify information i user is crossed, can be by view if user v accessed information iV, iNonzero value is entered as, such as 1;If user v is not
Information i was accessed, then by viewV, i0 is entered as, to realize ∑v∈VviewV, i·feedbackV, jWhat is calculated accessed to be all
Operation data caused by operation of the specify information i user to information to be pushed j.In addition, x and y is default primary data, its
Occurrence does not limit, for example, it may be x is 10, y 100.
For the coefficient of similarity sim between information i and information jI, j, information category is different, and its coefficient of similarity calculates can
With difference, if for example, information for application, coefficient of similarity simI, jCan be similar using the collaborative filtering between i and application j
Coefficient;If information is text, coefficient of similarity simI, jTf-idf coefficient of similarity that can be between text i and text j.
Specifically, the tf-idf similarity factors between any two text can be calculated in the following way:
Step S121:Obtain the words-frequency feature vector of text.
Specifically, can be by being segmented to text, the probability occurred using each participle is as the participle in the text
In weight, the weight is to segment the word frequency in article accordingly, and the words-frequency feature for obtaining text is vectorial.A for example, text
Text message in this is " typewriting most precisely, the input method of interface most individual character ", word segmentation result for " typewriting, precisely, it is interface, individual
Property, input method ", the probability that then each participle of statistics occurs, the probability occurred using each participle is as the participle to this article
This weight, then obtain word frequency difference of " typewriting, accurate, interface, personalization, input method " each participle in the text
For:" 0.2,0.2,0.2,0.2,0.2 ", the words-frequency feature vector of this article is:tfi={ typewriting:0.2, precisely:0.2, interface:
0.2, it is personalized:0.2, input method:0.2 }, wherein, tfiRepresent text i words-frequency feature vector.
Step S122:Calculate the inverse document frequency of different participles
Specifically, the calculation formula of inverse document frequency can be:
Wherein, idfwParticiple w inverse document frequency is represented, I represents the set of all texts in database,
isContainI, wRepresent whether participle w occurs in text i, can be represented to segment whether w occurs in text i with numerical value, such as
1 represents occur, and 0 represents do not occur.In addition, r is a constant, its concrete numerical value does not limit in the present embodiment, excellent
Choosing, r values are 1.
Step S123:Construct the tf-idf vectors of text.
Specifically, first, calculating the tf-idf values segmented in text, calculation formula can be:tfidfI, w=idfw·
tfI, w, wherein, tfidfI, wRepresent the tf-idf values of participle w in text i, idfwRepresent participle w inverse document frequency, tfI, wTable
Show the word frequency that w is segmented in text i.
The tf-idf values each segmented are calculated to all participles of article, you can the tf-idf vectors for obtaining article i are designated as
tfidfi, specifically, can be tfidfi={ tfidfI, 1, tfidfI, 2... }, represent each in text i tf-idf vectors
Coordinate is respectively the tf-idf values of first participle in text i, second tf-idf value segmented, the 3rd tf-idf segmented
Value etc., by that analogy.
Step S424:Calculate the coefficient of similarity between text
Specifically, the formula for calculating the coefficient of similarity between text can be simI, j=cos<tdidfi, tdidfj>, its
In, simI, jRepresent the coefficient of similarity between text i and text j, tdidfiRepresent text i tf-idf vectors, tdidfjTable
Show text j tf-idf vectors.
For any information to be pushed, according to its intensified learning similarity factor between each specify information, can obtain
Push coefficient corresponding to the information to be pushed.Specifically, can be by adding between each specify information and the information to be pushed
Intensified learning similarity factor, obtain push coefficient corresponding to the information to be pushed.Specifically, it can be counted by equation below
Calculate:
Wherein, predictlikeU, jRepresent to treat the push coefficient that push user u pushes information to be pushed j.viewU, iTable
Show user u to be pushed whether access information i, specifically, accessing information i, view as user u to be pushedU, iIt could be arranged to non-
Null value, such as 1;When push user has not visited information i, viewU, i0 is could be arranged to, is pushed with realizing by specify information with waiting
Intensified learning similarity factor between information calculates push coefficient, rfsimI, jIt is strong between expression information i and information to be pushed j
Chemistry practises similarity factor.
Therefore, in the present embodiment, the push coefficient for obtaining any one information to be pushed can be calculated.
Step S130:Coefficient is pushed according to corresponding to each information to be pushed, push letter is chosen from information to be pushed
Breath, is pushed to the user to be pushed.
In the information to be pushed pushed for user to be pushed, the push coefficient magnitude of each information may be different, its
In, the bigger information to be pushed of push coefficient, the interest of user corresponding to the client closer to the access first information, because
This, can choose the larger information to be pushed of push coefficient as pushed information, be pushed away to the client for accessing the first information
Send.
Wherein it is possible to the information to be pushed of the maximum predetermined number of push coefficient is chosen from all information to be pushed to visitor
Family end pushes.Specifically, information to be pushed can be arranged according to the size descending of push coefficient, predetermined number is waited to push away before selection
Breath deliver letters as the pushed information pushed to user to be pushed.The predetermined number can be set according to actual push demand
It is fixed, can be 200, certainly, its occurrence does not limit in the present embodiment in the present embodiment.
In the present embodiment, treat push user enter row information push time be not intended to limit, can user access visitor
Pushed during the end of family, predetermined time period can be spaced and once pushed to user, or other.
In summary, in information-pushing method provided in an embodiment of the present invention, specified based on what user to be pushed accessed
Information, and other accessed operation data of the user to information to be pushed of specify information, each information to be pushed pair of acquisition
The push coefficient answered, pushed information is chosen from information to be pushed according to coefficient to be pushed and carries out push to user to be pushed,
So that to user to be pushed push information closer to user to be pushed interest-degree.
Second embodiment
A kind of information push-delivery apparatus 200 is present embodiments provided, refers to Fig. 4, the device 200 includes:Acquisition of information mould
Block 210, the information accessed for obtaining user to be pushed, as specify information;Coefficients calculation block 220 is pushed, according to each
Intensified learning similarity factor between specify information and each information to be pushed, push coefficient corresponding to each information to be pushed is obtained,
Wherein, the intensified learning similarity factor between any specify information and any information to be pushed specifies letter according to this was accessed
The user of breath obtains to the operation data of the information to be pushed;Pushing module 230, for corresponding according to each information to be pushed
Push coefficient, pushed information is chosen from information to be pushed, is pushed to the user to be pushed.
Wherein, data obtaining module 210 can obtain the information that user to be pushed accessed in preset time period, as
Specify information;The information for the default number of branches that user to be pushed finally accesses either is obtained, as specify information.
Furthermore it is possible to user is previously provided with to operating parameter corresponding to the various operations of information, for any specified letter
Breath and any information to be pushed, described device can also include learning coefficient computing module, for will access the specify information
Operation data of the user to operating parameter corresponding to the operation of the information to be pushed as the user to the information to be pushed.
Specifically, the operating parameter set can include, in user in the accession page of specify information, being pushed away to the user
Information to be pushed is sent, the user does not click on the information to be pushed, and corresponding operating parameter is the first numerical value;In user to specifying
In the accession page of information, information to be pushed has been pushed to the user, the user clicks on and have accessed the information to be pushed, but to this
For information to be pushed without other behaviors, corresponding operating parameter is second value;In user in the accession page of specify information, to
The user has pushed information to be pushed, and the user clicks on and have accessed the information to be pushed, and generates it to the information to be pushed
His behavior, corresponding operating parameter is third value.
Further, can be used for for any specify information and any information to be pushed, learning coefficient computing module
According to all users for accessing the specify information are to operation data caused by the operation of the information to be pushed and this is to be specified
Coefficient of similarity between information and the information to be pushed, calculate the extensive chemical between the specify information and the information to be pushed
Practise similarity factor.
Specifically, the intensified learning that learning coefficient computing module is calculated between the specify information and the information to be pushed is similar
The formula of coefficient can be:
Wherein, the simI, jFor the coefficient of similarity between information i and information j, viewV, iRepresent whether user v accesses
Cross information i, V and represent all users, feedbackV, jRepresent operation data caused by operations of the user v to information to be pushed j, x
And y is default primary data.
Specifically, in the present embodiment, information can be text, the coefficient of similarity simI, jFor text i and text j
Between tf-idf coefficient of similarity.
In the present embodiment, can be by adding with each specify information with being somebody's turn to do for any information to be pushed, pushing module 230
Intensified learning similarity factor between information to be pushed, obtain push coefficient corresponding to the information to be pushed.
Specifically obtaining the formula of push coefficient corresponding to each information to be pushed can be:
Wherein, predictlikeU, jRepresent to treat the push coefficient that push user u pushes information to be pushed j, viewU, iTable
Show user u to be pushed whether access information i, rfsimI, jRepresent the similar system of intensified learning between information i and information to be pushed j
Number.
It should be noted that each embodiment in this specification is described by the way of progressive, each embodiment weight
Point explanation is all difference with other embodiment, between each embodiment identical similar part mutually referring to.
For device class embodiment, because it is substantially similar to embodiment of the method, so description is fairly simple, related part is joined
See the part explanation of embodiment of the method.
In several embodiments provided herein, it should be understood that disclosed apparatus and method, can also pass through
Other modes are realized.Device embodiment described above is only schematical, for example, flow chart and block diagram in accompanying drawing
Show the device of multiple embodiments according to the present invention, method and computer program product architectural framework in the cards,
Function and operation.At this point, each square frame in flow chart or block diagram can represent the one of a module, program segment or code
Part, a part for the module, program segment or code include one or more and are used to realize holding for defined logic function
Row instruction.It should also be noted that at some as in the implementation replaced, the function that is marked in square frame can also with different from
The order marked in accompanying drawing occurs.For example, two continuous square frames can essentially perform substantially in parallel, they are sometimes
It can perform in the opposite order, this is depending on involved function.It is it is also noted that every in block diagram and/or flow chart
The combination of individual square frame and block diagram and/or the square frame in flow chart, function or the special base of action as defined in performing can be used
Realize, or can be realized with the combination of specialized hardware and computer instruction in the system of hardware.
In addition, each functional module in each embodiment of the present invention can integrate to form an independent portion
Point or modules individualism, can also two or more modules be integrated to form an independent part.
If the function is realized in the form of software function module and is used as independent production marketing or in use, can be with
It is stored in a computer read/write memory medium.Based on such understanding, technical scheme is substantially in other words
The part to be contributed to prior art or the part of the technical scheme can be embodied in the form of software product, the meter
Calculation machine software product is stored in a storage medium, including some instructions are causing a computer equipment (can be
People's computer, server, or network equipment etc.) perform all or part of step of each embodiment methods described of the present invention.
And foregoing storage medium includes:USB flash disk, mobile hard disk, read-only storage (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), magnetic disc or CD etc. are various can be with the medium of store program codes.Need
Illustrate, herein, such as first and second or the like relational terms be used merely to by an entity or operation with
Another entity or operation make a distinction, and not necessarily require or imply between these entities or operation any this reality be present
The relation or order on border.Moreover, term " comprising ", "comprising" or its any other variant are intended to the bag of nonexcludability
Contain, so that process, method, article or equipment including a series of elements not only include those key elements, but also including
The other element being not expressly set out, or also include for this process, method, article or the intrinsic key element of equipment.
In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that including the key element
Process, method, other identical element also be present in article or equipment.
The preferred embodiments of the present invention are the foregoing is only, are not intended to limit the invention, for the skill of this area
For art personnel, the present invention can have various modifications and variations.Within the spirit and principles of the invention, that is made any repaiies
Change, equivalent substitution, improvement etc., should be included in the scope of the protection.It should be noted that:Similar label and letter exists
Similar terms is represented in following accompanying drawing, therefore, once being defined in a certain Xiang Yi accompanying drawing, is then not required in subsequent accompanying drawing
It is further defined and explained.
The foregoing is only a specific embodiment of the invention, but protection scope of the present invention is not limited thereto, any
Those familiar with the art the invention discloses technical scope in, change or replacement can be readily occurred in, should all be contained
Cover within protection scope of the present invention.Therefore, protection scope of the present invention described should be defined by scope of the claims.