CN107277115A - A kind of content delivery method and device - Google Patents

A kind of content delivery method and device Download PDF

Info

Publication number
CN107277115A
CN107277115A CN201710389954.XA CN201710389954A CN107277115A CN 107277115 A CN107277115 A CN 107277115A CN 201710389954 A CN201710389954 A CN 201710389954A CN 107277115 A CN107277115 A CN 107277115A
Authority
CN
China
Prior art keywords
content
targeted customer
historical
user
interest
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201710389954.XA
Other languages
Chinese (zh)
Inventor
王娜
王文君
陈昭男
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen University
Original Assignee
Shenzhen University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenzhen University filed Critical Shenzhen University
Priority to CN201710389954.XA priority Critical patent/CN107277115A/en
Publication of CN107277115A publication Critical patent/CN107277115A/en
Pending legal-status Critical Current

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/50Network services
    • H04L67/55Push-based network services
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/50Network services
    • H04L67/535Tracking the activity of the user

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer Hardware Design (AREA)
  • General Engineering & Computer Science (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

The present invention is applied to data analysis, and there is provided a kind of content delivery method and device with processing technology field.This method includes:The content for obtaining whole users checks historical data, calculate object content and the similarity of the historical content of targeted customer associated, obtain user's scoring of the historical content for the targeted customer that targeted customer couple associates with object content, calculate the time of the act weight that targeted customer checks the historical content of the targeted customer associated with object content, calculate interest-degree of the targeted customer to object content, targeted customer's interest-degree highest preset quantity object content is chosen, targeted customer is pushed to.Compared to prior art, this programme is being obtained during user pushes content, when calculating the interest-degree of user, introduces the time of the act weight of user's history content this parameter, make the statistics of user interest degree more accurate, and then make user's push content of acquisition more accurate.

Description

A kind of content delivery method and device
Technical field
The invention belongs to data analysis and processing technology field, more particularly to a kind of content delivery method and device.
Background technology
With gradually stepped into information epoch, the world today is in the environment of information huge explosion, while being faced with sternness Problem of information overload.Only in 2011, global metadata amount has just reached 1.8ZB, can be produced for each person every year equivalent to the whole world More than 200GB data, and this numeral also increasing year by year, according to conservative, it is expected that in the following years, data yield will Remain annual 50% growth rate.Nowadays, on major electric business, video playback platform, audio playing platform, user The data of magnanimity are all produced daily, therefore how to effectively utilize the data of user's generation is current Internet enterprises urgent need to resolve The problem of.Now, personalized commending system just arises at the historic moment as the means of data mining.Commending system refers to internet Website provides a user Item Information or suggestion, allows user to find oneself potential interest and demand and help user's candidate Product.
Collaborative filtering (item-based collaborative filtering) algorithm based on article is current industry Using most proposed algorithms.Either Amazon net, or Netflix, Hulu, YouTube etc., the basis of its proposed algorithm All it is the algorithm.Its advantage has:1. calculate simple;2. rationale for the recommendation can be concluded according to user's history behavior;3. user behavior History recommends efficiency higher.When its shortcoming mainly there are 1. number of articles much larger than user, article co-occurrence matrix cost is calculated It is too big;2. cold start-up problem is serious;3. selected seed article carries out calculating recommendation list, temporal information is not accounted for.
The content of the invention
Technical problem to be solved of the embodiment of the present invention is to provide a kind of content delivery method and device, it is intended to solve The problem of user interest degree calculates inaccurate in the prior art.
First aspect of the embodiment of the present invention provides a kind of content delivery method, and methods described includes:
The content for obtaining whole users checks historical data, and the content of the user checks that historical data includes the complete of user Portion's historical content and each historical content check time point that the historical content is the content that user checked;
The content associated with the historical content of targeted customer is defined as object content, the object content is calculated with closing The similarity of the historical content of the targeted customer of connection, it is described that the acquisition targeted customer couple associates with the object content User's scoring of the historical content of targeted customer, according to checking time point for each historical content, calculates the targeted customer Check the time of the act weight of the historical content of the targeted customer associated with the object content;
According to the similarity, user scoring and the time of the act weight, the targeted customer is calculated to described The interest-degree of object content;
Targeted customer's interest-degree highest preset quantity object content is chosen, the target is pushed to and uses Family.
Second aspect of the embodiment of the present invention provides a kind of content push device, and described device includes:
Acquisition module, checks historical data, the content of the user checks history number for obtaining the content of whole users According to checking time point for whole historical contents including user and each historical content, the historical content is that user checked Content;
Processing module, for the content associated with the historical content of targeted customer to be defined as into object content, calculates institute Object content and the similarity of the historical content of the targeted customer associated are stated, the targeted customer couple and the target is obtained User's scoring of the historical content of the targeted customer of relevance, according to checking time point for each historical content, meter Calculate the time of the act weight that the targeted customer checks the historical content of the targeted customer associated with the object content;
Computing module, for according to the similarity, user scoring and the time of the act weight, calculating the mesh Mark interest-degree of the user to the object content;
Pushing module, for choosing targeted customer's interest-degree highest preset quantity object content, is pushed To the targeted customer.
It was found from the embodiments of the present invention, the present invention checks historical data by obtaining the content of whole users, will be with The associated content of the historical content of targeted customer is defined as object content, calculates going through for object content and the targeted customer that associates The similarity of history content, obtains user's scoring of the historical content for the targeted customer that targeted customer couple associates with object content, root According to checking time point for each historical content, calculating targeted customer checks the historical content of the targeted customer associated with object content Time of the act weight, according to similarity, user's scoring and time of the act weight, calculates interest of the targeted customer to object content Degree, chooses targeted customer's interest-degree highest preset quantity object content, is pushed to targeted customer, compared to prior art, This programme, when calculating the interest-degree of user, introduces the behavior of user's history content during user's push content is obtained This parameter of time weighting, makes the statistics of user interest degree more accurate, and then makes user's push content of acquisition more accurate.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing There is the accompanying drawing used required in technology description to be briefly described, it should be apparent that, drawings in the following description are only this Some embodiments of invention, for those skilled in the art, without having to pay creative labor, can be with root Other accompanying drawings are obtained according to these accompanying drawings.
Accompanying drawing 1 is the implementation process schematic diagram for the content delivery method that first embodiment of the invention is provided;
Accompanying drawing 2 is the implementation process schematic diagram for the content delivery method that second embodiment of the invention is provided;
Accompanying drawing 3 is the structural representation for the content push device that third embodiment of the invention is provided;
Accompanying drawing 4 is the structural representation for the content push device that fourth embodiment of the invention is provided;
Accompanying drawing 5 is the user behavior history matrix that second embodiment of the invention is provided;
Accompanying drawing 6 is the schematic diagram that the object content interest-degree that second embodiment of the invention is provided is calculated.
Embodiment
To enable goal of the invention, feature, the advantage of the embodiment of the present invention more obvious and understandable, below in conjunction with Accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that retouched The embodiment stated is only a part of embodiment of the invention, and not all embodiments.Based on the embodiment in the present invention, this area The every other embodiment that technical staff is obtained under the premise of creative work is not made, belongs to the model that the present invention is protected Enclose.
Refer to accompanying drawing 1, the implementation process signal for the content delivery method that accompanying drawing 1 provides for first embodiment of the invention Figure, this method can apply in terminal device.As shown in Figure 1, this method is mainly included the following steps that:
S101, the content of the whole users of acquisition check historical data;
Wherein, the content of user checks that historical data includes whole historical contents and when the checking of each historical content of user Between point.Further, the historical content is the content that user checked, i.e., pass through the end before the user under terminal device records The content that end equipment was checked.The historical content can be, but not limited to include:Video, audio, news or commodity on network.Look into The mode seen includes clicking on the link of the historical content.
S102, the content associated with the historical content of targeted customer be defined as to object content, calculate object content with The similarity of the historical content of the targeted customer of association, obtains the history for the targeted customer that targeted customer couple associates with object content User's scoring of content, according to checking time point for each historical content, calculates targeted customer and checks the mesh associated with object content Mark the time of the act weight of the historical content of user;
If there is user had both checked a certain content, and the historical content of targeted customer, then it is assumed that the content is and target The associated content of the historical content of user, object content is defined as by the content.
When targeted customer scores the historical content no user of targeted customer, it is 1 to give tacit consent to user scoring.
S103, according to similarity, user's scoring and time of the act weight, calculate targeted customer to the interest of object content Degree;
S104, selection targeted customer's interest-degree highest preset quantity object content, are pushed to targeted customer.
It should be understood that preset quantity herein can as needed be configured, change.
Content delivery method provided in an embodiment of the present invention, historical data is checked by the content for obtaining whole users, will The content associated with the historical content of targeted customer is defined as object content, calculates object content and the targeted customer's that associates The similarity of historical content, obtains user's scoring of the historical content for the targeted customer that targeted customer couple associates with object content, According to checking time point for each historical content, the historical content that targeted customer checks the targeted customer associated with object content is calculated Time of the act weight, according to similarity, user's scoring and time of the act weight, calculate targeted customer to the interest of object content Degree, chooses targeted customer's interest-degree highest preset quantity object content, is pushed to targeted customer, compared to prior art, This programme, when calculating the interest-degree of user, introduces the behavior of user's history content during user's push content is obtained This parameter of time weighting, makes the statistics of user interest degree more accurate, and then makes user's push content of acquisition more accurate.
Refer to accompanying drawing 2, the implementation process signal for the content delivery method that accompanying drawing 2 provides for second embodiment of the invention Figure, this method can apply in terminal device.As shown in Figure 2, this method is mainly included the following steps that:
S201, the content of the whole users of acquisition check historical data;
Wherein, the content of user checks that historical data includes whole historical contents and when the checking of each historical content of user Between point.Further, the historical content is the content that user checked, i.e., pass through the end before the user under terminal device records The content that end equipment was checked.The historical content can be, but not limited to include:Video, audio, news or commodity on network.Look into The mode seen includes clicking on the link of the historical content.
S202, the content associated with the historical content of targeted customer be defined as object content;
If there is user had both checked a certain content, and the historical content of targeted customer, then it is assumed that the content is and target The associated content of the historical content of user, object content is defined as by the content.
S203, historical data checked according to the content of whole users of acquisition, set up user behavior history matrix;
As shown in figure 5, Fig. 5 is the user behavior history matrix that terminal device is set up, wherein A, B, C, D, E are use Family, a, b, c, d, e are historical content.
S204, according to user behavior history matrix and formulaCalculate object content and associate The similarity of the historical content of targeted customer;
Wherein wijFor object content and the similarity of the historical content of targeted customer associated, N (i) is to look into whole users The historical content i of the targeted customer associated with object content number of users is seen, N (j) is to check in target in whole users Hold j number of users, N (i) ∩ N (j) is while checked i and j number of users.By what is counted in user behavior history matrix Data bring formula intoIn, calculate object content similar with the historical content of targeted customer that associates Degree.So that user behavior history matrix is Fig. 5 as an example, it is assumed that object content is a, then
S205, the user for the historical content for obtaining the targeted customer that targeted customer couple associates with object content score;
Wherein, when targeted customer scores the historical content no user of targeted customer, it is 1 to give tacit consent to user scoring.
S206, according to formulaCalculate targeted customer and check going through for the targeted customer associated with object content The time of the act weight of history content;
δ is the interest attenuation factor, and the interest attenuation factor can be adjusted as needed.tuiChecked for targeted customer u The logical reach of the newest behavior of historical content i distance objective users of the targeted customer associated with object content, targeted customer is most New behavior is that targeted customer checks row of the time point away from the nearest content of current point in time is checked in the historical content of targeted customer For.Targeted customer u checks that i's checks that time point and targeted customer u check the checking between time point of the newest behavior of targeted customer Targeted customer historical content number it is more when, tuiValue it is bigger, tuiFor nonnegative integer.
S207, according to formulaCalculate interest of the targeted customer to object content Degree;
PujFor interest-degrees of the targeted customer u to object content j, N (u) is the set of targeted customer u whole historical contents, S (j, K) is the set in targeted customer u historical content with object content j similarities K historical content of highest, wijFor mesh Mark content j and targeted customer u historical content i similarity, ruiFor use of the targeted customer u to the historical content i of targeted customer Family is scored, luiThe historical content i of targeted customer time of the act weight is checked for targeted customer u.
S208, selection targeted customer's interest-degree highest preset quantity object content, are pushed to targeted customer.
It should be understood that preset quantity herein can as needed be configured, change.
By taking Fig. 6 as an example, it is assumed that whole historical contents that targeted customer u is checked by the order of the time of checking from morning to night are distinguished For:A, B, C, D, now D is the newest behavior of targeted customer, then can make tuD=0, tuC=1, tuB=2, tuA=3.User couple A, B, C, D scoring are respectively 0.7,0.6,0.5 and 0.8, and interest attenuation factor delta=10 are taken here, K=3 is taken, with A similarities 3 videos of highest are a, b, c, w respectivelyaA=0.9, wbA=0.8, wcA=0.7;Distinguish with 3 videos of B similarities highest It is d, e, f, wdB=0.7, weB=0.6, wfB=0.5;It is f, g, h, w respectively with 3 videos of C similarities highestfC=0.6, wgC=0.5, whC=0.4;It is i, j, k, w respectively with 3 videos of D similarities highestiD=0.8, wjD=0.7, wkD=0.6, Wherein f be both with one in 3 videos of B similarities highest, be again and one in 3 videos of C similarities highest.
User u is P to a interest-degreeua, PuaIt is calculated as follows:
User u is P to f interest-degreeuf, PufIt is calculated as follows:
According to calculating, Pua=0.467, Pub=0.415, Puc=0.363, Pud=0.344, Pue=0.295, Puf= 0.517、Pug=0.226, Puh=0.181, Pui=0.64, Puj=0.56, Puk=0.48.
Obtained interest-degree descending arrangement will be calculated, and takes preceding TopN object content to recommend user.If for example, taken When Top5 object content recommends user, recommendation list is [i, j, f, k, a].
Content delivery method provided in an embodiment of the present invention, historical data is checked by the content for obtaining whole users, will The content associated with the historical content of targeted customer is defined as object content, calculates object content and the targeted customer's that associates The similarity of historical content, obtains user's scoring of the historical content for the targeted customer that targeted customer couple associates with object content, According to checking time point for each historical content, the historical content that targeted customer checks the targeted customer associated with object content is calculated Time of the act weight, according to similarity, user's scoring and time of the act weight, calculate targeted customer to the interest of object content Degree, chooses targeted customer's interest-degree highest preset quantity object content, is pushed to targeted customer, compared to prior art, This programme, when calculating the interest-degree of user, introduces the behavior of user's history content during user's push content is obtained This parameter of time weighting, makes the statistics of user interest degree more accurate, and then makes user's push content of acquisition more accurate.
Accompanying drawing 3 is referred to, accompanying drawing 3 is the structural representation for the content push device that third embodiment of the invention is provided, and is It is easy to explanation, illustrate only the part related to the embodiment of the present invention.The content push device of the example of accompanying drawing 3 can be foregoing The executive agent for the content delivery method that first embodiment is provided, it can be a work(in terminal device or terminal device Can module.The content push device of the example of accompanying drawing 3, mainly includes:Acquisition module 301, processing module 302, computing module 303 and Pushing module 304.Each functional module describes in detail as follows:
Acquisition module 301, checks historical data, the content of user checks historical data for obtaining the content of whole users Whole historical contents and each historical content including user check time point that historical content is the content that user checked.
Processing module 302, for the content associated with the historical content of targeted customer to be defined as into object content, is calculated Object content and the similarity of the historical content of targeted customer associated, obtain the target that targeted customer couple associates with object content User's scoring of the historical content of user, according to checking time point for each historical content, calculate targeted customer check with target Hold the time of the act weight of the historical content of the targeted customer of association.
Computing module 303, for according to similarity, user's scoring and time of the act weight, calculating targeted customer to target The interest-degree of content.
Pushing module 304, for choosing targeted customer's interest-degree highest preset quantity object content, is pushed to target User.
The detailed process of the respective function of above-mentioned each Implement of Function Module, the content for referring to aforementioned first embodiment offer is pushed away The related content of delivery method, here is omitted.
Content push device provided in an embodiment of the present invention, historical data is checked by the content for obtaining whole users, will The content associated with the historical content of targeted customer is defined as object content, calculates object content and the targeted customer's that associates The similarity of historical content, obtains user's scoring of the historical content for the targeted customer that targeted customer couple associates with object content, According to checking time point for each historical content, the historical content that targeted customer checks the targeted customer associated with object content is calculated Time of the act weight, according to similarity, user's scoring and time of the act weight, calculate targeted customer to the interest of object content Degree, chooses targeted customer's interest-degree highest preset quantity object content, is pushed to targeted customer, compared to prior art, This programme, when calculating the interest-degree of user, introduces the behavior of user's history content during user's push content is obtained This parameter of time weighting, makes the statistics of user interest degree more accurate, and then makes user's push content of acquisition more accurate.
Accompanying drawing 4 is referred to, accompanying drawing 4 is the structural representation for the content push device that fourth embodiment of the invention is provided, and is It is easy to explanation, illustrate only the part related to the embodiment of the present invention.The content push device of the example of accompanying drawing 4 can be foregoing The executive agent for the content delivery method that second embodiment is provided, it can be a work(in terminal device or terminal device Can module.The content push device of the example of accompanying drawing 4, mainly includes:Acquisition module 401, processing module 402, computing module 403 and Pushing module 404.Each functional module describes in detail as follows:
Acquisition module 401, checks historical data, the content of user checks historical data for obtaining the content of whole users Whole historical contents and each historical content including user check time point that historical content is the content that user checked.
Processing module 402, for the content associated with the historical content of targeted customer to be defined as into object content, according to The content of whole users of acquisition checks historical data, sets up user behavior history matrix.
Processing module 402, is additionally operable to according to user behavior history matrix and formulaCalculate mesh Mark content and the similarity of the historical content of targeted customer associated, wherein wijFor going through for object content and the targeted customer that associates The similarity of history content, N (i) is the number of users for checking the historical content i of the targeted customer associated with object content, N (j) To check object content j number of users, N (i) ∩ N (j) is while checked i and j number of users.
Processing module 402, is additionally operable to obtain the historical content for the targeted customer that targeted customer couple associates with object content User scores, according to formulaTargeted customer is calculated to check in the history of the targeted customer associated with object content The time of the act weight of appearance, wherein δ is the interest attenuation factor, tuiCheck that the target associated with object content is used for targeted customer u The logical reach of the newest behavior of historical content i distance objective users at family, the newest behavior of targeted customer is that targeted customer checks mesh Behavior of the time point away from the nearest content of current point in time is checked in the historical content for marking user.
Computing module 403, for according to formulaTargeted customer is calculated to mesh Mark the interest-degree of content, wherein PujFor interest-degrees of the targeted customer u to object content j, N (u) is targeted customer u whole history With object content j similarities K historical content of highest in the set of content, the historical content that S (j, K) is targeted customer u Set, wijFor object content j and targeted customer u historical content i similarity, ruiTargeted customer is gone through for targeted customer u History content i user's scoring, luiThe historical content i of targeted customer time of the act weight is checked for targeted customer u.
Pushing module 404, for choosing targeted customer's interest-degree highest preset quantity object content, is pushed to target User.
The detailed process of the respective function of above-mentioned each Implement of Function Module, the content for referring to aforementioned second embodiment offer is pushed away The related content of delivery method, here is omitted.
Content push device provided in an embodiment of the present invention, historical data is checked by the content for obtaining whole users, will The content associated with the historical content of targeted customer is defined as object content, calculates object content and the targeted customer's that associates The similarity of historical content, obtains user's scoring of the historical content for the targeted customer that targeted customer couple associates with object content, According to checking time point for each historical content, the historical content that targeted customer checks the targeted customer associated with object content is calculated Time of the act weight, according to similarity, user's scoring and time of the act weight, calculate targeted customer to the interest of object content Degree, chooses targeted customer's interest-degree highest preset quantity object content, is pushed to targeted customer, compared to prior art, This programme, when calculating the interest-degree of user, introduces the behavior of user's history content during user's push content is obtained This parameter of time weighting, makes the statistics of user interest degree more accurate, and then makes user's push content of acquisition more accurate.
It should be noted that for foregoing each method embodiment, for simplicity description, therefore it is all expressed as a series of Combination of actions, but those skilled in the art should know, the present invention is not limited by described sequence of movement because According to the present invention, some steps can use other orders or carry out simultaneously.Secondly, those skilled in the art should also know Know, embodiment described in this description belongs to preferred embodiment, and involved action and module might not all be this hairs Necessary to bright.
In the above-described embodiments, the description to each embodiment all emphasizes particularly on different fields, and does not have the portion being described in detail in some embodiment Point, it may refer to the associated description of other embodiments.
It is above description to content delivery method provided by the present invention, device, for those skilled in the art, according to According to the thought of the embodiment of the present invention, it will change in specific embodiments and applications, to sum up, in this specification Appearance should not be construed as limiting the invention.

Claims (10)

1. a kind of content delivery method, it is characterised in that methods described includes:
The content for obtaining whole users checks historical data, and the content of the user checks that whole of the historical data including user is gone through History content and each historical content check time point that the historical content is the content that user checked;
The content associated with the historical content of targeted customer be defined as object content, the object content is calculated and associates The similarity of the historical content of the targeted customer, obtains the target that the targeted customer couple associates with the object content User's scoring of the historical content of user, according to checking time point for each historical content, calculates the targeted customer and checks The time of the act weight of the historical content of the targeted customer associated with the object content;
According to the similarity, user scoring and the time of the act weight, the targeted customer is calculated to the target The interest-degree of content;
Targeted customer's interest-degree highest preset quantity object content is chosen, the targeted customer is pushed to.
2. content delivery method as claimed in claim 1, it is characterised in that described to be commented according to the similarity, the user Divide and the time of the act weight, calculate interest-degree of the targeted customer to the object content, including:
According to formulaCalculate interest of the targeted customer to the object content Degree, wherein PujFor interest-degrees of the targeted customer u to the object content j, N (u) is whole history of the targeted customer u Gone through in the set of content, the historical content that S (j, K) is the targeted customer u with the object content j similarity highests K The set of history content, wijFor the object content j and historical content i of targeted customer u similarity, ruiFor the mesh Mark user u to score to the historical content i of targeted customer user, luiThe targeted customer is checked for the targeted customer u Historical content i time of the act weight.
3. content delivery method as claimed in claim 2, it is characterised in that the calculating object content and the institute associated The similarity of the historical content of targeted customer is stated, including:
Historical data is checked according to the content of whole users of acquisition, user behavior history matrix is set up;
According to the user behavior history matrix and formulaCalculate the object content and associate The similarity of the historical content of the targeted customer, wherein wijFor going through for the object content and the targeted customer that associates The similarity of history content, N (i) is the user for checking the historical content i of the targeted customer associated with the object content Quantity, N (j) is the number of users for checking the object content j, and N (i) ∩ N (j) is while checked i and j number of users Amount.
4. content delivery method as claimed in claim 2, it is characterised in that described according to each historical content when checking Between point, calculate the time of the act that the targeted customer checks the historical content of the targeted customer associated with the object content Weight, including:
According to formulaCalculate the targeted customer and check the targeted customer's that is associated with the object content The time of the act weight of historical content, wherein δ is the interest attenuation factor, tuiFor the targeted customer u check with the target Hold the logical reach of the newest behavior of historical content i distance objective users of the targeted customer of association, the targeted customer is most New behavior is that the targeted customer checks and checks that time point is nearest away from current point in time in the historical content of the targeted customer The behavior of content.
5. the content delivery method as described in any one of claim 2 to 4, it is characterised in that when the targeted customer u is to described During the historical content i no users scoring of targeted customer, r is setuiValue be 1.
6. a kind of content push device, it is characterised in that described device includes:
Acquisition module, checks historical data, the content of the user checks historical data bag for obtaining the content of whole users Include whole historical contents of user and checking time point for each historical content, the historical content be user checked it is interior Hold;
Processing module, for the content associated with the historical content of targeted customer to be defined as into object content, calculates the mesh The similarity of mark content and the historical content of targeted customer associate, the acquisition targeted customer couple and the object content User's scoring of the historical content of the targeted customer of association, according to checking time point for each historical content, calculates institute State the time of the act weight that targeted customer checks the historical content of the targeted customer associated with the object content;
Computing module, for according to the similarity, user scoring and the time of the act weight, calculating the target and using Interest-degree of the family to the object content;
Pushing module, for choosing targeted customer's interest-degree highest preset quantity object content, is pushed to institute State targeted customer.
7. content push device as claimed in claim 6, it is characterised in that
The computing module, specifically for according to formulaCalculate the targeted customer To the interest-degree of the object content, wherein PujFor interest-degrees of the targeted customer u to the object content j, N (u) is institute State the set of targeted customer u whole historical contents, S (j, K) in the historical content of the targeted customer u with the target Hold the set of j similarities K historical content of highest, wijFor the object content j and targeted customer u historical content i Similarity, ruiThe historical content i of targeted customer user is scored for the targeted customer u, luiUsed for the target Family u checks the historical content i of targeted customer time of the act weight.
8. content push device as claimed in claim 7, it is characterised in that
The processing module, is additionally operable to check historical data according to the content of whole users of acquisition, sets up user behavior history Matrix;
According to the user behavior history matrix and formulaCalculate the object content and associate The similarity of the historical content of the targeted customer, wherein wijFor the object content and the history of the targeted customer associated The similarity of content, N (i) is the number of users for checking the historical content i of the targeted customer associated with the object content Amount, N (j) is the number of users for checking the object content j, and N (i) ∩ N (j) is while checked i and j number of users.
9. content push device as claimed in claim 7, it is characterised in that
The processing module, is additionally operable to according to formulaThe targeted customer is calculated to check and the object content The time of the act weight of the historical content of the targeted customer of association, wherein δ is the interest attenuation factor, tuiUsed for the target Family u checks the logic of the newest behavior of historical content i distance objective users of the targeted customer associated with the object content Distance, the newest behavior of targeted customer is that the targeted customer checks in the historical content of the targeted customer and checks time point Behavior away from the nearest content of current point in time.
10. the content push device as described in any one of claim 7 to 9, it is characterised in that when the targeted customer u is to institute When stating the historical content i no users scoring of targeted customer, r is setuiValue be 1.
CN201710389954.XA 2017-05-27 2017-05-27 A kind of content delivery method and device Pending CN107277115A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710389954.XA CN107277115A (en) 2017-05-27 2017-05-27 A kind of content delivery method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710389954.XA CN107277115A (en) 2017-05-27 2017-05-27 A kind of content delivery method and device

Publications (1)

Publication Number Publication Date
CN107277115A true CN107277115A (en) 2017-10-20

Family

ID=60064754

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710389954.XA Pending CN107277115A (en) 2017-05-27 2017-05-27 A kind of content delivery method and device

Country Status (1)

Country Link
CN (1) CN107277115A (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018218403A1 (en) * 2017-05-27 2018-12-06 深圳大学 Content pushing method and device
WO2020037931A1 (en) * 2018-08-20 2020-02-27 平安科技(深圳)有限公司 Item recommendation method and apparatus, computer device and storage medium
CN111046229A (en) * 2018-10-11 2020-04-21 广东阿里影业云智软件有限公司 Information pushing method and server side equipment
CN111062734A (en) * 2018-10-16 2020-04-24 北京字节跳动网络技术有限公司 Method and device for reordering release information, electronic equipment and readable medium
CN111949860A (en) * 2019-05-15 2020-11-17 北京字节跳动网络技术有限公司 Method and apparatus for generating a relevance determination model
WO2020257993A1 (en) * 2019-06-24 2020-12-30 深圳市欢太科技有限公司 Content pushing method and apparatus, server, and storage medium
WO2021047119A1 (en) * 2019-09-12 2021-03-18 广东浪潮大数据研究有限公司 Information pushing method and system, electronic device and storage medium

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7979445B2 (en) * 2003-10-13 2011-07-12 Amazon Technologies, Inc. Processes for assessing user affinities for particular item categories of a hierarchical browse structure
CN104391849A (en) * 2014-06-30 2015-03-04 浙江大学苏州工业技术研究院 Collaborative filtering recommendation method for integrating time contextual information

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7979445B2 (en) * 2003-10-13 2011-07-12 Amazon Technologies, Inc. Processes for assessing user affinities for particular item categories of a hierarchical browse structure
CN104391849A (en) * 2014-06-30 2015-03-04 浙江大学苏州工业技术研究院 Collaborative filtering recommendation method for integrating time contextual information

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
印桂生等: "遗忘曲线的协同过滤推荐模型", 《哈尔滨工程大学学报》 *
史艳翠,戴浩男,石和平,汪圣洁,杨硕珩,钟惠军: "一种基于时间戳的新闻推荐模型", 《计算机应用与软件》 *
项亮: "《推荐系统实践》第一章、第二章、第五章", 《推荐系统实践》 *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018218403A1 (en) * 2017-05-27 2018-12-06 深圳大学 Content pushing method and device
WO2020037931A1 (en) * 2018-08-20 2020-02-27 平安科技(深圳)有限公司 Item recommendation method and apparatus, computer device and storage medium
CN111046229A (en) * 2018-10-11 2020-04-21 广东阿里影业云智软件有限公司 Information pushing method and server side equipment
CN111062734A (en) * 2018-10-16 2020-04-24 北京字节跳动网络技术有限公司 Method and device for reordering release information, electronic equipment and readable medium
CN111062734B (en) * 2018-10-16 2023-08-25 北京字节跳动网络技术有限公司 Method and device for reordering delivery information, electronic equipment and readable medium
CN111949860A (en) * 2019-05-15 2020-11-17 北京字节跳动网络技术有限公司 Method and apparatus for generating a relevance determination model
CN111949860B (en) * 2019-05-15 2022-02-08 北京字节跳动网络技术有限公司 Method and apparatus for generating a relevance determination model
WO2020257993A1 (en) * 2019-06-24 2020-12-30 深圳市欢太科技有限公司 Content pushing method and apparatus, server, and storage medium
WO2021047119A1 (en) * 2019-09-12 2021-03-18 广东浪潮大数据研究有限公司 Information pushing method and system, electronic device and storage medium

Similar Documents

Publication Publication Date Title
CN107277115A (en) A kind of content delivery method and device
JP5802745B2 (en) Intelligent navigation method, apparatus and system
CN104424291B (en) The method and device that a kind of pair of search result is ranked up
US20150278355A1 (en) Temporal context aware query entity intent
CN105159932B (en) A kind of data retrieval engine and ordering system and method
CN108874812B (en) Data processing method, server and computer storage medium
US20100293170A1 (en) Social network message categorization systems and methods
CN106250513A (en) A kind of event personalization sorting technique based on event modeling and system
EP2570935A1 (en) Method for personalized meta-search and application terminal thereof
CN104239373B (en) Add tagged method and device for document
CN108170692A (en) A kind of focus incident information processing method and device
CN103258025B (en) Generate the method for co-occurrence keyword, the method that association search word is provided and system
CN103838756A (en) Method and device for determining pushed information
KR20150036117A (en) Query expansion
CN105378730A (en) Social media content analysis and output
CN105224529A (en) A kind of personalized recommendation method based on user browsing behavior and device
CN105023178B (en) A kind of electronic commerce recommending method based on ontology
CN109460519B (en) Browsing object recommendation method and device, storage medium and server
CN106850750A (en) A kind of method and apparatus of real time propelling movement information
CN107911448A (en) Content pushing method and device
CN112818230B (en) Content recommendation method, device, electronic equipment and storage medium
CN103064903A (en) Method and device for searching images
CN105426550A (en) Collaborative filtering tag recommendation method and system based on user quality model
US10127322B2 (en) Efficient retrieval of fresh internet content
CN102855245A (en) Image similarity determining method and image similarity determining equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20171020