CN102999588A - Method and system for recommending multimedia applications - Google Patents
Method and system for recommending multimedia applications Download PDFInfo
- Publication number
- CN102999588A CN102999588A CN2012104620790A CN201210462079A CN102999588A CN 102999588 A CN102999588 A CN 102999588A CN 2012104620790 A CN2012104620790 A CN 2012104620790A CN 201210462079 A CN201210462079 A CN 201210462079A CN 102999588 A CN102999588 A CN 102999588A
- Authority
- CN
- China
- Prior art keywords
- multimedia application
- user
- recommendation list
- recommendation
- initial
- 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
Links
Images
Abstract
The invention discloses a method and a system for recommending multimedia applications. The method comprises the steps of: calculating recommendation parameters corresponding to users according to historic behavior characteristics of the users using the multimedia applications, wherein the recommendation parameters represent the preference degree of the certain user for the at least one multimedia application; calculating the similarity between the at least two multimedia applications according to the current recommendation parameters; generating initial multimedia application recommendation lists for the users according to the similarity; and obtaining the initial multimedia application recommendation lists when receiving the recommendation requests of the current users so as to show the initial multimedia application recommendation lists at the clients of the current users. The method and system disclosed by the embodiment of the invention have the advantages of being capable of supplying different and personalized multimedia application recommendation services for the different users and increasing the accuracy and success rate of multimedia application recommendation.
Description
Technical field
The present invention relates to the network data processing field, particularly relate to a kind of recommend method and system of multimedia application.
Background technology
Multimedia application is a kind of application that the multimedia services such as voice or video are provided to the user.Multimedia application comprises channel, chatroom etc., channel, it is the voice/video content service product that provides on a kind of internet platform, for example support team's voice communication instrument---the QQ Talk room that multi-person speech exchanges, the user enters channel can receive voice communication, the services such as discussion, education, amusement of playing that channel provides.Video chat chambers 9158 of the chatroom also is the voice/video content service product that provides on a kind of internet platform, the voice-enabled chat net---9 chat chamber, Video chat website---9158.com etc.
Prior art generally adopts dual mode when realizing that multimedia application is recommended, the first adopts the mode of the title keyword search of channel/chatroom.It is its required channel/chatroom of keyword search of user inputting channel/chatroom.The second adopts the unified mode of recommending in channel/chatroom, and for example the channel of YY voice is ranked and the activity centre, and wherein, the mode of ranking list is a kind of of non-personalized recommendation, and the mathematical statistics of the basic overall situation provides the rank on popularity or the flow.For example click " channel seniority among brothers and sisters " button in the YY voice software, server can return the most front channel of popularity seniority among brothers and sisters, the channel of the perhaps VHI after the manual intervention, and classification seniority among brothers and sisters.And the similar a kind of marketing way of recommendation in activity centre, in advance advance notice has channel or the chatroom of relevant occasion.For example: click " activity centre " in the YY voice software, server returns the channel ads of holding in the recent period large-scale activity.
The inventor finds that in research process first kind of way of the prior art belongs to the user search mode, needs the user to input relevant information, such as: the title key words of channel/chatroom etc. need the user to know in advance the relevant information of these channel/chatrooms.For the uncomprehending channel of user or chatroom, then can't recommend the user.And the second way of the prior art, although do not need the user to understand in advance channel and input relevant information.But this recommendation or based on original popularity seniority among brothers and sisters, or based on walking advertisement.For all users, the channel of recommendation or chatroom all are the same, do not have to form the personalized recommendation for the user.
Therefore, this dual mode all can be so that the recommendation accuracy of prior art when carrying out the multimedia application recommendation be not high, not only make prior art Matthew effect when multimedia application is recommended, occur, and then affect the interactive efficiency of user and multimedia application platform, if and the user hits for inappropriate multimedia application is overdue, just make the multimedia application platform trigger some unnecessary user responses, also affect the utilization of the system resource of multimedia application platform.
Summary of the invention
The application's technical matters to be solved provides a kind of recommend method of multimedia application, in order to solve the personalized recommendation that does not have in the prior art to form for the user, avoid when multimedia application is recommended, occurring Matthew effect, and the interactive efficiency of the multimedia application platform that causes is low, affects the problem of utilization of the system resource of multimedia application platform.
The application also provides a kind of recommendation apparatus of multimedia application, in order to guarantee said method implementation and application in practice.
In order to address the above problem, the application discloses a kind of recommend method of multimedia application, and the method comprises:
Multimedia application recommendation list generative process:
According to the user's who uses described multimedia application historical behavior feature, calculate recommended parameter corresponding to described user, described recommended parameter represents that certain user is for the preference degree of certain at least one multimedia application;
According to the similarity between at least two multimedia application of described current recommended parameter calculating;
Generate user's initial multimedia application recommendation list according to described similarity;
The multimedia application recommendation process:
When receiving active user's recommendation request, obtain described initial multimedia application and recommend row to show with the client described active user.
Preferably, described user's according to using described multimedia application historical behavior feature is calculated recommended parameter corresponding to described each user, comprising:
Obtain eigenwert and the power of each historical behavior feature of the user who uses described multimedia application
Heavy;
The described eigenwert of foundation and the described user of weight calculation be recommended parameter one to one.
Preferably, the described current recommended parameter of described foundation is calculated the similarity between at least two multimedia application, comprising:
The user is defined as the vector of described any multimedia application for the recommended parameter of any multimedia application;
According to the vector of each multimedia application, the mode of employing Euclidean distance is calculated the similarity between at least two multimedia application.
Preferably, also comprise:
Receive described active user for the click data of the multimedia application in the described initial multimedia application recommendation list;
Calculate the hit rate of described initial multimedia application recommendation list according to described click data.
Preferably, also comprise:
Judge that whether described hit rate satisfies default recommendation condition, if not, then adjusts described active user corresponding to the recommended parameter of at least one multimedia application according to described hit rate.
Preferably, describedly obtain described initial multimedia application and recommend row to show with the client described active user, comprising:
Obtain described active user's initial multimedia application recommendation list;
According to the default principle of optimality described initial multimedia application recommendation list is optimized;
The client that multimedia application recommendation list after optimizing is sent to described active user is showed.
Preferably, describedly according to the default principle of optimality described initial multimedia application recommendation list is optimized, comprises:
Filter the multimedia application that service no longer is provided in the described initial multimedia application recommendation list;
Whether the number of multimedia application satisfies predetermined threshold value in the multimedia application recommendation list after judge filtering, if so, then carries out the step that client that multimedia application recommendation list behind described will the optimization is sent to described active user is showed;
If not, then obtaining the default multimedia application of supplying tabulates, supply the multimedia application of supplying of extracting default number the multimedia application tabulation according to default decimation rule from described, with supplying in the multimedia application recommendation list after multimedia application is added into described filtration as the multimedia application recommendation list after the optimization of described default number.
Disclosed herein as well is a kind of commending system of multimedia application, comprise multimedia application recommendation list generating apparatus and multimedia application recommendation apparatus:
Described multimedia application recommendation list generating apparatus comprises:
Calculated recommendation parameter unit is used for the historical behavior feature according to the user who uses described multimedia application, calculates recommended parameter corresponding to described user, and described recommended parameter represents that certain user is for the preference degree of certain at least one multimedia application;
Calculate the similarity unit, be used for according to the similarity between at least two multimedia application of described current recommended parameter calculating;
Generate list cell, generate user's initial multimedia application recommendation list according to described similarity;
Described multimedia application recommendation apparatus is used for obtaining described initial multimedia application and recommends row to show with the client described active user.
Preferably, described calculated recommendation parameter unit comprises:
Obtain eigenwert and weight module, for eigenwert and the weight of each historical behavior feature of obtaining the user who uses described multimedia application;
The calculated recommendation parameter module is used for one to one recommended parameter of the described eigenwert of foundation and the described user of weight calculation.
Preferably, described calculating similarity unit comprises:
Determine vector module, be used for the user is defined as for the recommended parameter of any multimedia application the vector of described any multimedia application;
Calculate the similarity module, be used for the vector according to each multimedia application, the mode of employing Euclidean distance is calculated the similarity between at least two multimedia application.
Preferably, also comprise:
Receive and click data cell, be used for receiving described active user for the click data of the multimedia application of described initial multimedia application recommendation list;
Calculate the hit rate unit, be used for calculating according to described click data the hit rate of described initial multimedia application recommendation list.
Preferably, also comprise:
Judging unit is used for judging whether described hit rate satisfies default recommendation condition;
Adjustment unit is used for result at described judging unit and is adjusting described active user corresponding to the recommended parameter of at least one multimedia application according to described hit rate in the no situation.
Preferably, described multimedia application recommendation apparatus comprises:
Obtain initial multimedia application recommendation list unit, be used for when receiving active user's recommendation request, obtaining described active user's initial multimedia application recommendation list;
Optimize the unit, be used for according to the default principle of optimality described initial multimedia application recommendation list being optimized;
Transmitting element, the multimedia application recommendation list after being used for optimizing are sent to described active user's client to be showed.
Preferably, described optimization unit comprises:
Filtering module is used for filtering the multimedia application that described initial multimedia application recommendation list no longer provides service;
Judge module is used for judging whether the number of the multimedia application recommendation list multimedia application after filtering satisfies predetermined threshold value;
Trigger module, the result who is used at described judge module is in the situation that is, triggers described sending module;
Obtain and supply list block, be used for result at described judge module and be in the no situation, obtain the default multimedia application tabulation of supplying;
Abstraction module is used for supplying the multimedia application tabulation according to the multimedia application of supplying of the default number of default decimation rule extraction from described;
Add module, be used for will described default number the multimedia application of supplying be added into multimedia application recommendation list after the described filtration multimedia application recommendation list after as optimization.
With respect to prior art, beneficial effect of the present invention is: the technical scheme that adopts present embodiment, at first, do not need the user that multimedia application is understood in advance, do not need the user to input any information yet, secondly, the application can the quantification user to the preference degree of multimedia application as recommended parameter, calculated different recommended parameters for different users, thereby can provide for different user the recommendation service of differentiation and Extraordinary multimedia application, avoid when multimedia application is recommended, occurring Matthew effect, development long-tail product has also promoted accuracy and success ratio that multimedia application is recommended, and then promotes the interactive efficiency of user and multimedia application platform, also avoid the multimedia application platform to trigger the phenomenon that some unnecessary users respond, promoted the utilization factor of the system resource of multimedia application platform.
Description of drawings
In order to be illustrated more clearly in the technical scheme in the embodiment of the present application, the accompanying drawing of required use was done to introduce simply during the below will describe embodiment, apparently, accompanying drawing in the following describes only is some embodiment of the application, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 is the process flow diagram of the recommend method embodiment 1 of the disclosed multimedia application of the application;
Fig. 2 is the process flow diagram of step 101 in the application's embodiment of the method 1;
Fig. 3 is the process flow diagram of step 102 in the application's embodiment of the method 1;
Fig. 4 is the process flow diagram of step 104 in the application's embodiment of the method 1;
Fig. 5 is the process flow diagram of step 402 in the application's embodiment of the method 1;
Fig. 6 is the process flow diagram of the recommend method embodiment 2 of the disclosed multimedia application of the application;
Fig. 7 is the structured flowchart of the commending system embodiment 1 of the disclosed a kind of multimedia application of the application;
Fig. 8 is the structured flowchart that calculates recommended parameter unit 701 in the application's system embodiment 1;
Fig. 9 is the structured flowchart that calculates similarity unit 702 in the application's system embodiment 1;
Figure 10 is the structured flowchart of multimedia application recommendation apparatus 71 in the application's system embodiment 1;
Figure 11 is the structured flowchart of optimizing unit 1002 in the application's system embodiment 1;
Figure 12 is the structured flowchart of the commending system embodiment 2 of the disclosed a kind of multimedia application of the application.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present application, the technical scheme in the embodiment of the present application is clearly and completely described, obviously, described embodiment only is the application's part embodiment, rather than whole embodiment.Based on the embodiment among the application, those of ordinary skills are not making the every other embodiment that obtains under the creative work prerequisite, all belong to the scope of the application's protection.
The embodiment of the present application realizes based on the principle of collaborative filtering.Collaborative filtering (CollaborativeFiltering, CF) refer generally in mass users, to excavate out like sub-fraction and some users' the grade comparing class, in collaborative filtering, these users that excavated become neighbours, and the catalogue that other things of then liking according to them are organized into an ordering returns to the active user as recommendation results.
With reference to figure 1, show the process flow diagram of the recommend method embodiment 1 of a kind of multimedia application of the application, in the present embodiment, step 101~step 103 is multimedia application recommendation list generative processes, step 104 then is the multimedia application recommendation process, these two processes can be separate in actual applications, and multimedia application recommendation list generative process can periodically initiatively be carried out, and the multimedia application recommendation process can carry out when user's triggered multimedia is used recommendation again.Present embodiment specifically can comprise:
Step 101: according to the user's who uses multimedia application historical behavior feature, calculate recommended parameter corresponding to described user, described recommended parameter represents that certain user is for the preference degree of certain at least one multimedia application.
In the present embodiment, needing each historical behavior feature according to the user who uses multimedia application, calculate the user for the preference degree of certain at least one multimedia application, namely is recommended parameter corresponding to each user.Wherein, the historical behavior feature can comprise: the duration that the user stops in multimedia application (for example channel or chatroom), the multimedia application of user's collection, the role hierarchy of user in multimedia application, the information such as the channel that the user resides or chatroom.But when implementing the application, those skilled in the art can select in the historical behavior feature any one or a few to come the calculated recommendation parameter, perhaps also can adopt other users' historical behavior feature to come the calculated recommendation parameter again.
Concrete, with reference to shown in Figure 2, Fig. 2 is the process flow diagram of step 101, step 101 can comprise in the specific implementation:
Step 201: eigenwert and the weight of obtaining each historical behavior feature of the user who uses described multimedia application;
At first, the data of the user's that collects historical behavior feature are carried out normalization, obtain the eigenwert of each historical behavior feature.Normalized target is to form the user to the recommended parameter of multimedia application.Wherein, the span of recommended parameter can be arranged in 0 to 1 interval.
Be understandable that, when different users' historical behavior feature is carried out value, all eigenwerts dropped in 0 to 1 interval.Wherein, this historical behavior feature of duration that in multimedia application (for example channel or chatroom), stops for the user, usually have a long way to go between maximal value and the minimum value, and skewness, the time long data that stops for the user presents the long-tail characteristics, so can carry out noise reduction process in multimedia application, thereby reduce the impact that the two poles of the earth data are brought to value, long data trends towards normal distribution when making.
When weight is arranged, those skilled in the art can be a user's the identical weight of all historical behavior feature default settings when implementing the application, also can to each different historical behavior feature different weights be set according to the actual requirements, the weight sum that only needs to guarantee same user is 1 to get final product.
Step 202: the described eigenwert of foundation and the described user of weight calculation be recommended parameter one to one.
In the embodiment of the present application, after the eigenwert and weight of each historical behavior feature that obtains the user, can utilize computing formula (1) to calculate a user to the recommended parameter of some multimedia application:
Recommended parameter=∑ (user behavior feature value * weight) (1)
As seen, recommended parameter is relevant with user and multimedia application, and all there is a recommended parameter of unique correspondence in each user to each multimedia application, thus the output format of recommended parameter can for:
<User?ID><Channel?ID><Preference?Value>
Wherein, " UserID " is used for user of unique identification, and " Channel ID " is used for multimedia application of unique identification, " Preference Value " is recommended parameter, represent certain user to the preference degree of certain multimedia application, span is in [0,1] interval.
After step 101 calculated recommendation parameter, the recommended parameter that calculates can be stored, because calculated recommendation parameter and follow-up to recommend multimedia application to the user may not be that the time is upper closely continuous, so need first recommended parameter to be stored, so that again triggered multimedia recommendation of using during the follow-up recommendation request that receives the user.Need to prove, because not necessarily carry out recommendation process after the calculated recommendation parameter at once, so after step 101, execution in step 102 at once not necessarily, get final product so only need step 102 after step 101, to carry out, whether after step 101, carry out and need not to limit step 102 at once.
Step 102: according to the similarity between at least two multimedia application of described current recommended parameter calculating.
Before introducing the application and calculating the step of similarity, understand collaborative filtering mode based on article in order to make things convenient for those skilled in the art, first the collaborative filtering mode based on article is described in detail at this.Collaborative filtering based on article is based on user's article similar to the preference of article, then recommends similar article according to user's historical preference to it.From the angle of calculating, exactly all users are calculated similarity between the article to the preference of certain article as a vector, after obtaining the similar article of certain article, predict that according to user's historical preference the active user does not also represent the article of preference, calculates the item lists of an ordering as recommendation.For example, for article A, according to all users' historical preference, like the user of article A all to like article C, just can draw the article A conclusion more similar with article C, and user C likes article A, can infer that so user C may also like article C, just recommends user C with article C.
After getting access to recommended parameter, use the Collaborative Filtering Recommendation Algorithm based on article, calculate the similarity of multimedia application (corresponding to article), thereby can realize calculating the multimedia application recommendation list by similarity.
Concrete, with reference to shown in Figure 3, Fig. 3 is the process flow diagram of step 102, step 102 can comprise in the specific implementation:
Step 301: the vector that the user is defined as described any multimedia application for the recommended parameter of any multimedia application;
In this application for the calculating of the similarity of multimedia application, be based on vector (Vector), and in the two-dimensional matrix of the recommended parameter of user-multimedia application, namely be that all users are calculated similarity between the channel to the recommended parameter of certain multimedia application as a vector, namely be the distance of calculating two vectors, the similarity that these two vectors of the nearlyer explanation of distance that is to say these two multimedia application is larger.
Step 302: according to the vector of each multimedia application, the mode of employing Euclidean distance is calculated the similarity between at least two multimedia application.
After determining vector, can use in this application Euclidean distance to calculate similarity between any two multimedia application, the computing formula of the Euclidean distance between two multimedia application is as follows:
Wherein, d (x, y) namely is the Euclidean distance between multimedia application x and the y, and N is user's number.
Similarity between two multimedia application then adopts computing formula (3) to calculate:
Wherein, sim (x, y) namely is the similarity between multimedia application x and the y.Wherein, the numerical value of similarity is larger, illustrates that two distances between the multimedia application are less, that is to say more similar.
Step 103: the initial multimedia application recommendation list that generates each user according to described similarity.
After the similarity between any two multimedia application is determined, can recommend some other multimedia application similar to this multimedia application to it according to certain multimedia application of active user's login, namely be front several multimedia application the most similar to this certain multimedia application of user's login, it is formed initial multimedia application recommendation list; Perhaps when the user searches multimedia application, recommend some other multimedia application similar to the multimedia application of its search to it, and these other multimedia application namely are front several multimedia application the most similar to the multimedia application of family search, also it are formed initial multimedia application recommendation list.For example, the form of this initial multimedia application recommendation list can for<User ID〉<Channel ID1,<Channel ID2 〉,<Channel ID3 ...<Channel ID n〉}, namely be to recommend n multimedia application for a user.
Wherein, need to prove, need not calculate again in the recommendation request that receives the user for the calculating of recommended parameter in the embodiment of the present application and obtain, generation for the calculating of preference data and initial multimedia application recommendation list can be periodically initiatively to calculate, when receiving the multimedia application recommendation request that certain user triggers, obtain again the initial multimedia application recommendation list of its correspondence follow-up.
Step 104: when receiving active user's recommendation request, obtain described initial multimedia application and recommend row to show with the client described active user.
The recommendation request that receives the user in this step can be understood as a kind of trigger condition, in actual scene, can be to detect website or the platform that the user has logined certain multimedia application, for example the user has entered the QQTalk chatroom, then begin to carry out the recommendation process of follow-up multimedia application, also can be to detect the user to have submitted the request of searching multimedia application to, also can carry out the recommendation process of follow-up multimedia application this moment.
Concrete, in actual applications, the flow process of described step 104 can be with reference to shown in Figure 4, and step 104 specifically can comprise:
Step 401: when receiving active user's recommendation request, obtain described active user's initial multimedia application recommendation list.
Initial multimedia application recommendation list namely is the multimedia application recommendation list that generates in step step 103 in this step.
Step 402: described initial multimedia application recommendation list is optimized according to the default principle of optimality.
Need to prove, may there be the situation such as seal and stop in multimedia application in the initial multimedia application recommendation list that only obtains according to dimension of similarity, therefore, also need in the embodiment of the present application initial multimedia application recommendation list is optimized according to the default principle of optimality and recommend again the active user.
Concrete, with reference to shown in Figure 5, Fig. 5 is the process flow diagram of step 402, step 402 can comprise in the specific implementation:
Step 501: filter the multimedia application that service no longer is provided in the described initial multimedia application recommendation list.
Because multimedia application active sometimes intersexuality, so excessively low channel or the chatroom of liveness can be filtered when recommending multimedia application, for example, do not have the user to use in the Preset Time, perhaps be less than user's use of certain preset number etc. in the Preset Time.Simultaneously because therefore the calculating of multimedia application recommendation and recommended parameter and non real-time carrying out, initially the situation that some multimedia application are sealed and stopped may occur in the multimedia application recommendation list, so also must filter these multimedia application.
Step 502: whether the number of judging multimedia application in the multimedia application recommendation list after filtering satisfies predetermined threshold value, if not, then enters step 503;
Filtering out in the initial multimedia application recommendation list no longer provides after the multimedia application of service,
Whether the number of judging again multimedia application in the multimedia application recommendation list after filtering satisfies predetermined threshold value, for example, default recommendation number is 10, and through after the filtration of step 501, only surplus 8 multimedia application in the multimedia application recommendation list so just need to enter step 503 and carry out replenishing of multimedia application.And if the number of multimedia application satisfies predetermined threshold value in the multimedia application recommendation list after filtering, then can enter step 403, namely be that the client that the multimedia application recommendation list after optimizing is sent to described active user is showed.
Step 503: obtain the default multimedia application of supplying and tabulate.
For the situation of the multimedia application number deficiency that may occur recommending after filtering, need to supply, at first obtain the default multimedia application of supplying and tabulate.Need to prove, supplying the multimedia application tabulation can generate in advance, for example, select the more much higher media application of some popularities to add and supply the multimedia application tabulation, certainly, when the multimedia application tabulation was supplied in generation, the admissible dimension of those skilled in the art can have a lot, gives an example no longer one by one at this.This is supplied the multimedia application tabulation and can be stored in the storer, obtains from storer when needed.Simultaneously, can also safeguard supplying multimedia application tabulator labour movement battalion, upgrade or replacement etc.
Step 504: extract the multimedia application of supplying of presetting number according to default decimation rule from described supplying the multimedia application tabulation.
From supply the multimedia application tabulation, extract again the multimedia application of supplying of default number according to default decimation rule, wherein, default decimation rule can be for example: the preferential extraction recommends number of times less than 500, be in the of zero from number of times, secondly extract and to recommend number of times more than or equal to 500, from beginning of shooting straight etc.Concrete default decimation rule also can independently be arranged by those skilled in the art.In addition, the not much higher media application of some clicking rates can no longer extract it, perhaps it is removed from supply the multimedia application tabulation.Wherein, for example, if need to show 10 multimedia application, and only have 8 in the tabulation of the multimedia application after filtering, this step just needs to extract two multimedia application so.
Step 505: with supplying in the multimedia application recommendation list after multimedia application is added into described filtration as the multimedia application recommendation list after optimizing of described default number.
Supplying in the multimedia application recommendation list after multimedia application is added into filtration as the multimedia application recommendation list after optimizing of the default number that will be drawn into again is in order to follow-uply return to the active user with it.
Then return Fig. 4, enter step 403: the multimedia application recommendation list after will optimizing is sent to described active user's client and shows.
Multimedia application recommendation list after will optimizing at last is sent to described active user's client and shows, with each multimedia application in user terminal displays multimedia application recommendation list, can also when the request of family when at every turn refreshing the multimedia application recommendation list used in reception, rearrange this multimedia application recommendation list.
The application specifically can select the Hadoop platform to do the system architecture of realization when implementing, utilize the Distributed Calculation cluster that erects excellent performance that the Hadoop platform can be rapid, cheap, can select Mahout as the architecture of data recommendation engine simultaneously.Hadoop is open source code concurrent operation programming tool and Distributed Calculation and the storage system that Apache Software Foundation is researched and developed, and is similar with the concept of MapReduce and Google file system; Apache Mahout is project of increasing income of Apache Software Foundation, and it comes from Lucene, is structured on the Hadoop, pays close attention to the efficient realization of the machine learning classic algorithm on the mass data.Wherein as the Hadoop of distributed calculating and storage platform, also other platforms be can be replaced with in actual use, data storage and calculating carried out such as database platform.And the Mahout that uses makees the proposed algorithm instrument, also can use other Data Mining Tools such as SAS to substitute in actual the use, even realize voluntarily related algorithm.Simultaneously, the application has used collaborative filtering as proposed algorithm, also can use other algorithms in actual the use, recommends to calculate such as cluster, and effect may be not so good as collaborative filtering certainly.
Adopt the technical scheme of present embodiment, at first, do not need the user that multimedia application is understood in advance, do not need the user to input any information yet, secondly, the application can the quantification user to the preference degree of multimedia application as recommended parameter, calculated different recommended parameters for different users, thereby can provide for different user the recommendation service of differentiation and Extraordinary multimedia application, accuracy and success ratio that multimedia application is recommended have been promoted, solved prior art in the problem of carrying out occurring when multimedia application is recommended Matthew effect, and then also promote the interactive efficiency of user and multimedia application platform, also avoid the multimedia application platform to trigger the phenomenon that some unnecessary users respond, promoted the utilization factor of the system resource of multimedia application platform.
With reference to shown in Figure 6, show the process flow diagram of the recommend method embodiment 2 of a kind of multimedia application of the application, after the step 101 in implementing embodiment 1~step 104, present embodiment can also may further comprise the steps:
Step 601: receive described active user for the click data of the multimedia application in the initial multimedia application recommendation list.
Present embodiment according to the click data of user for the multimedia application in the initial multimedia application recommendation list, is also adjusted user's recommended parameter as different from Example 1.At first, receiving the active user for the click data of the multimedia application in the initial multimedia application recommendation list, namely is whether the user has clicked the multimedia application of recommending.
Step 602: the hit rate of calculating described initial multimedia application recommendation list according to described click data.
After receiving user's click data, can set up mathematical model, user and multimedia application before and after recommending are analyzed, estimate the recommendation effect of multimedia application.Wherein, the important indicator of evaluation is hit rate, and hit rate adopts computing formula (4) to calculate:
Number of clicks/recommendation multimedia application number of times * 100% (4)
Wherein, " number of clicks " refers to that the user is to the number of clicks of some multimedia application of recommendation; " recommend multimedia application number of times " refers to the number of times that the user is recommended in some multimedia application, the recommendation number of times of the multimedia application that in other words to be exactly the user see in client.
Step 603: judge that whether described hit rate satisfies default recommendation condition, if not, then enters step 604.
In this step, can see the hit rate that calculates whether in a default numerical range,
If, illustrate that then this hit rate satisfies default recommendation condition, and if just do not illustrating that this hit rate does not satisfy default recommendation condition.
Specifically when implementing, also can adopt the mode of grouping comparison evaluation to examine hit rate, namely one group is not used the recommendation list (at random) of present techniques scheme and one group of recommendation list of using the present techniques scheme to obtain out to carry out the contrast of hit rate, if the latter's hit rate meets default recommendation condition than the former high explanation, otherwise does not then meet.
Step 604: adjust described active user corresponding to the recommended parameter of at least one multimedia application according to described hit rate.
And when hit rate did not satisfy default recommendation condition, the active user who calculates in can set-up procedure 101 was corresponding to the recommended parameter of at least one multimedia application.For example, can attempt eigenwert and the weight of certain historical behavior feature that will this current user heightens, see again whether the hit rate that the recommended parameter after the adjustment calculates satisfies default recommendation condition, if satisfy, illustrate that then this adjustment tallies with the actual situation, if and satisfied, then past opposite direction is adjusted eigenwert and the weight of certain historical behavior feature of this current user, until finally obtain meeting the hit rate of default recommendation condition.
In the present embodiment, when initial multimedia application recommendation list does not all meet default recommendation condition, can also guarantee by the adjustment to recommended parameter the accuracy of initial multimedia application recommendation list, thereby guaranteed to provide for different user the realization of the recommendation service of differentiation and Extraordinary multimedia application, solved prior art in the problem of carrying out occurring when multimedia application is recommended Matthew effect, accuracy and success ratio that multimedia application is recommended have also been promoted, and then the interactive efficiency of lifting user and multimedia application platform, also avoid the multimedia application platform to trigger the phenomenon that some unnecessary users respond, promoted the utilization factor of the system resource of multimedia application platform.
For aforesaid each embodiment of the method, for simple description, so it all is expressed as a series of combination of actions, but those skilled in the art should know, the application is not subjected to the restriction of described sequence of movement, because according to the application, some step can adopt other orders or carry out simultaneously.Secondly, those skilled in the art also should know, the embodiment described in the instructions all belongs to preferred embodiment, and related action and module might not be that the application is necessary.
Corresponding with the method that the recommend method embodiment 1 of a kind of multimedia application of above-mentioned the application provides, referring to Fig. 7, the application also provides a kind of commending system embodiment 1 of multimedia application, in the present embodiment, this system can comprise multimedia application recommendation list generating apparatus 70 and multimedia application recommendation apparatus 71, wherein, described multimedia application recommendation list generating apparatus 70 comprises:
Calculated recommendation parameter unit 701 is used for the historical behavior feature according to the user who uses described multimedia application, calculates recommended parameter corresponding to described user, and described recommended parameter represents that certain user is for the preference degree of at least one multimedia application.
With reference to shown in Figure 8, be the structured flowchart of described calculated recommendation parameter unit 701, calculated recommendation parameter unit 701 specifically can comprise:
Obtain eigenwert and weight module 801, for eigenwert and the weight of each historical behavior feature of obtaining the user who uses described multimedia application;
Calculated recommendation parameter module 802 is used for one to one recommended parameter of the described eigenwert of foundation and the described user of weight calculation.
Calculate similarity unit 702, be used for according to the similarity between at least two multimedia application of described current recommended parameter calculating.
With reference to shown in Figure 9, be the structured flowchart of described calculating similarity unit 702, calculate similarity unit 702 and specifically can comprise:
Determine vector module 901, be used for the user is defined as for the recommended parameter of any multimedia application the vector of described any multimedia application;
Calculate similarity module 902, be used for the vector according to each multimedia application, the mode of employing Euclidean distance is calculated the similarity between at least two multimedia application.
Generate list cell 703, generate user's initial multimedia application recommendation list according to described similarity.
Described multimedia application recommendation apparatus 71 is used for when receiving active user's recommendation request, obtains described initial multimedia application and recommends row to show with the client described active user.
With reference to shown in Figure 10, be described multimedia application recommendation apparatus 71 structured flowchart in actual applications, specifically can comprise:
Obtain initial multimedia application recommendation list unit 1001, be used for when receiving active user's recommendation request, obtaining described active user's initial multimedia application recommendation list.
With reference to shown in Figure 11, be the structured flowchart of described optimization unit 1002, described optimization unit 1002 specifically can comprise:
Obtain and supply list block 1104, be used for result at described judge module and be in the no situation, obtain the default multimedia application tabulation of supplying;
Add module 1106, be used for will described default number the multimedia application of supplying be added into multimedia application recommendation list after the described filtration multimedia application recommendation list after as optimization.
Transmitting element 1003, the multimedia application recommendation list after being used for optimizing are sent to described active user's client to be showed.
Adopt the technical scheme of present embodiment, at first, do not need the user that multimedia application is understood in advance, do not need the user to input any information yet, secondly, the application can the quantification user to the preference degree of multimedia application as recommended parameter, calculated different recommended parameters for different users, thereby can provide for different user the recommendation service of differentiation and Extraordinary multimedia application, accuracy and success ratio that multimedia application is recommended have been promoted, solved prior art in the problem of carrying out occurring when multimedia application is recommended Matthew effect, and then the interactive efficiency of lifting user and multimedia application platform, also avoid the multimedia application platform to trigger the phenomenon that some unnecessary users respond, promoted the utilization factor of the system resource of multimedia application platform.
Corresponding with the method that the recommend method embodiment 2 of a kind of multimedia application of above-mentioned the application provides, referring to Figure 12, the application also provides a kind of commending system embodiment 2 of multimedia application, in the present embodiment, except described multimedia application recommendation list generating apparatus 70 and multimedia application recommendation apparatus 71, this system can also comprise:
Receive to click data cell 1201, be used for receiving described active user for the click data of the multimedia application of described initial multimedia application recommendation list;
Calculate hit rate unit 1202, be used for calculating according to described click data the hit rate of described initial multimedia application recommendation list.
Judging unit 1203 is used for judging whether described hit rate satisfies default recommendation condition.
In the present embodiment, when initial multimedia application recommendation list does not all meet default recommendation condition, can also guarantee by the adjustment to recommended parameter the accuracy of initial multimedia application recommendation list, thereby guaranteed to provide for different user the realization of the recommendation service of differentiation and Extraordinary multimedia application, solved prior art in the problem of carrying out occurring when multimedia application is recommended Matthew effect, accuracy and success ratio that multimedia application is recommended have also been promoted, and then the interactive efficiency of lifting user and multimedia application platform, also avoid the multimedia application platform to trigger the phenomenon that some unnecessary users respond, promoted the utilization factor of the system resource of multimedia application platform.
Need to prove that each embodiment in this instructions all adopts the mode of going forward one by one to describe, what each embodiment stressed is and the difference of other embodiment that identical similar part is mutually referring to getting final product between each embodiment.For device class embodiment because itself and embodiment of the method basic simlarity, so describe fairly simple, relevant part gets final product referring to the part explanation of embodiment of the method.
At last, also need to prove, in this article, relational terms such as the first and second grades only is used for an entity or operation are made a distinction with another entity or operation, and not necessarily requires or hint and have the relation of any this reality or sequentially between these entities or the operation.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thereby not only comprise those key elements so that comprise process, method, article or the equipment of a series of key elements, but also comprise other key elements of clearly not listing, or also be included as the intrinsic key element of this process, method, article or equipment.Do not having in the situation of more restrictions, the key element that is limited by statement " comprising ... ", and be not precluded within process, method, article or the equipment that comprises described key element and also have other identical element.
More than recommend method and the system of a kind of multimedia application that the application is provided be described in detail, used specific case herein the application's principle and embodiment are set forth, the explanation of above embodiment just is used for helping to understand the application's method and core concept thereof; Simultaneously, for one of ordinary skill in the art, the thought according to the application all will change in specific embodiments and applications, and in sum, this description should not be construed as the restriction to the application.
Claims (14)
1. the recommend method of a multimedia application is characterized in that, the method comprises:
Multimedia application recommendation list generative process:
According to the user's who uses described multimedia application historical behavior feature, calculate recommended parameter corresponding to described user, described recommended parameter represents that certain user is for the preference degree of certain at least one multimedia application;
According to the similarity between at least two multimedia application of described current recommended parameter calculating;
Generate user's initial multimedia application recommendation list according to described similarity;
The multimedia application recommendation process:
When receiving active user's recommendation request, obtain described initial multimedia application and recommend row to show with the client described active user.
2. method according to claim 1 is characterized in that, described user's according to using described multimedia application historical behavior feature is calculated recommended parameter corresponding to described each user, comprising:
Obtain eigenwert and the weight of each historical behavior feature of the user who uses described multimedia application;
The described eigenwert of foundation and the described user of weight calculation be recommended parameter one to one.
3. method according to claim 1 is characterized in that, the described current recommended parameter of described foundation is calculated the similarity between at least two multimedia application, comprising:
The user is defined as the vector of described any multimedia application for the recommended parameter of any multimedia application;
According to the vector of each multimedia application, the mode of employing Euclidean distance is calculated the similarity between at least two multimedia application.
4. method according to claim 1 is characterized in that, also comprises:
Receive described active user for the click data of the multimedia application in the described initial multimedia application recommendation list;
Calculate the hit rate of described initial multimedia application recommendation list according to described click data.
5. method according to claim 4 is characterized in that, also comprises:
Judge that whether described hit rate satisfies default recommendation condition, if not, then adjusts described active user corresponding to the recommended parameter of at least one multimedia application according to described hit rate.
6. method according to claim 1 is characterized in that, describedly obtains described initial multimedia application and recommends row to show with the client described active user, comprising:
Obtain described active user's initial multimedia application recommendation list;
According to the default principle of optimality described initial multimedia application recommendation list is optimized;
The client that multimedia application recommendation list after optimizing is sent to described active user is showed.
7. method according to claim 6 is characterized in that, describedly according to the default principle of optimality described initial multimedia application recommendation list is optimized, and comprising:
Filter the multimedia application that service no longer is provided in the described initial multimedia application recommendation list;
Whether the number of multimedia application satisfies predetermined threshold value in the multimedia application recommendation list after judge filtering, if so, then carries out the step that client that multimedia application recommendation list behind described will the optimization is sent to described active user is showed;
If not, then obtaining the default multimedia application of supplying tabulates, supply the multimedia application of supplying of extracting default number the multimedia application tabulation according to default decimation rule from described, with supplying in the multimedia application recommendation list after multimedia application is added into described filtration as initial multimedia application recommendation list of described default number.
8. the commending system of a multimedia application is characterized in that, comprises multimedia application recommendation list generating apparatus and multimedia application recommendation apparatus:
Described multimedia application recommendation list generating apparatus comprises:
Calculated recommendation parameter unit is used for the historical behavior feature according to the user who uses described multimedia application, calculates recommended parameter corresponding to described user, and described recommended parameter represents that certain user is for the preference degree of at least one multimedia application;
Calculate the similarity unit, be used for according to the similarity between at least two multimedia application of described current recommended parameter calculating;
Generate list cell, generate user's initial multimedia application recommendation list according to described similarity;
Described multimedia application recommendation apparatus is used for when receiving active user's recommendation request, obtains described initial multimedia application and recommends row to show with the client described active user.
9. system according to claim 8 is characterized in that, described calculated recommendation parameter unit comprises:
Obtain eigenwert and weight module, for eigenwert and the weight of each historical behavior feature of obtaining the user who uses described multimedia application;
The calculated recommendation parameter module is used for one to one recommended parameter of the described eigenwert of foundation and the described user of weight calculation.
10. system according to claim 8 is characterized in that, described calculating similarity unit comprises:
Determine vector module, be used for the user is defined as for the recommended parameter of any multimedia application the vector of described any multimedia application;
Calculate the similarity module, be used for the vector according to each multimedia application, the mode of employing Euclidean distance is calculated the similarity between at least two multimedia application.
11. system according to claim 8 is characterized in that, also comprises:
Receive and click data cell, be used for receiving described active user for the click data of the multimedia application of described initial multimedia application recommendation list;
Calculate the hit rate unit, be used for calculating according to described click data the hit rate of described initial multimedia application recommendation list.
12. system according to claim 11 is characterized in that, also comprises:
Judging unit is used for judging whether described hit rate satisfies default recommendation condition;
Adjustment unit is used for result at described judging unit and is adjusting described active user corresponding to the recommended parameter of at least one multimedia application according to described hit rate in the no situation.
13. system according to claim 8 is characterized in that, described multimedia application recommendation apparatus comprises:
Obtain initial multimedia application recommendation list unit, be used for when receiving active user's recommendation request, obtaining described active user's initial multimedia application recommendation list;
Optimize the unit, be used for according to the default principle of optimality described initial multimedia application recommendation list being optimized;
Transmitting element, the multimedia application recommendation list after being used for optimizing are sent to described active user's client to be showed.
14. system according to claim 13 is characterized in that, described optimization unit comprises:
Filtering module is used for filtering the multimedia application that described initial multimedia application recommendation list no longer provides service;
Judge module is used for judging whether the number of the multimedia application recommendation list multimedia application after filtering satisfies predetermined threshold value;
Trigger module, the result who is used at described judge module is in the situation that is, triggers described sending module;
Obtain and supply list block, be used for result at described judge module and be in the no situation, obtain the default multimedia application tabulation of supplying;
Abstraction module is used for supplying the multimedia application tabulation according to the multimedia application of supplying of the default number of default decimation rule extraction from described;
Add module, be used for will described default number the multimedia application of supplying be added into multimedia application recommendation list after the described filtration multimedia application recommendation list after as optimization.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2012104620790A CN102999588A (en) | 2012-11-15 | 2012-11-15 | Method and system for recommending multimedia applications |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2012104620790A CN102999588A (en) | 2012-11-15 | 2012-11-15 | Method and system for recommending multimedia applications |
Publications (1)
Publication Number | Publication Date |
---|---|
CN102999588A true CN102999588A (en) | 2013-03-27 |
Family
ID=47928156
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2012104620790A Pending CN102999588A (en) | 2012-11-15 | 2012-11-15 | Method and system for recommending multimedia applications |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102999588A (en) |
Cited By (39)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103297853A (en) * | 2013-06-07 | 2013-09-11 | 华东师范大学 | IPTV (internet protocol television) program recommendation method based on context recognition for multiple users |
CN103309984A (en) * | 2013-06-17 | 2013-09-18 | 腾讯科技(深圳)有限公司 | Data processing method and device |
CN103593434A (en) * | 2013-11-12 | 2014-02-19 | 北京奇虎科技有限公司 | Application recommendation method and device and server equipment |
CN103685498A (en) * | 2013-12-06 | 2014-03-26 | 北京奇虎科技有限公司 | Wallpaper loading method and browser device |
CN103702226A (en) * | 2013-12-31 | 2014-04-02 | 广州华多网络科技有限公司 | Displaying method and device of direct broadcasting client end channel information |
CN103955484A (en) * | 2014-04-09 | 2014-07-30 | 微梦创科网络科技(中国)有限公司 | Method and system for estimating tendency of user to network social tools |
CN104065981A (en) * | 2014-06-20 | 2014-09-24 | 海信集团有限公司 | Method and device for recommending videos |
CN104182449A (en) * | 2013-05-20 | 2014-12-03 | Tcl集团股份有限公司 | System and method for personalized video recommendation based on user interests modeling |
CN104243590A (en) * | 2014-09-19 | 2014-12-24 | 广州华多网络科技有限公司 | Resource object recommendation method and device |
CN104603753A (en) * | 2014-03-19 | 2015-05-06 | 华为技术有限公司 | Method, system and server for recommending application |
CN104636406A (en) * | 2013-11-15 | 2015-05-20 | 华为技术有限公司 | Method and device for pushing information according to user behaviors |
CN104735150A (en) * | 2015-03-27 | 2015-06-24 | 努比亚技术有限公司 | Message pushing method and device |
WO2015131803A1 (en) * | 2014-03-06 | 2015-09-11 | Tencent Technology (Shenzhen) Company Limited | Application recommending method and system |
CN104915354A (en) * | 2014-03-12 | 2015-09-16 | 深圳市腾讯计算机系统有限公司 | Multimedia file pushing method and device |
CN105095305A (en) * | 2014-05-20 | 2015-11-25 | 深圳市腾讯计算机系统有限公司 | Method and device for generating personalized page |
CN105307023A (en) * | 2015-10-30 | 2016-02-03 | 北京奇艺世纪科技有限公司 | Application recommending method and apparatus, and video recommending method and apparatus |
CN105740268A (en) * | 2014-12-10 | 2016-07-06 | 阿里巴巴集团控股有限公司 | Information pushing method and apparatus |
CN105787053A (en) * | 2016-02-26 | 2016-07-20 | 维沃移动通信有限公司 | Application pushing method and terminal |
CN105868248A (en) * | 2015-12-15 | 2016-08-17 | 乐视网信息技术(北京)股份有限公司 | Media recommendation method and device |
CN105930423A (en) * | 2016-04-18 | 2016-09-07 | 乐视控股(北京)有限公司 | Multimedia similarity determination method and apparatus as well as multimedia recommendation method |
CN103678518B (en) * | 2013-11-28 | 2017-02-15 | 北京邮电大学 | Method and device for adjusting recommendation lists |
CN106484766A (en) * | 2016-09-07 | 2017-03-08 | 北京百度网讯科技有限公司 | Searching method based on artificial intelligence and device |
CN106951418A (en) * | 2016-01-07 | 2017-07-14 | 广州启法信息科技有限公司 | A kind of personalized recommendation technology of business web site |
CN107045688A (en) * | 2016-02-05 | 2017-08-15 | 上海行邑信息科技有限公司 | A kind of control method and device of accurate identification target terminal |
CN107306355A (en) * | 2016-04-20 | 2017-10-31 | 华为技术有限公司 | A kind of content recommendation method and server |
CN107330747A (en) * | 2017-05-16 | 2017-11-07 | 深圳和而泰智能家居科技有限公司 | Beauty appliance gear recommends method, beauty appliance and storage medium |
CN107526753A (en) * | 2016-07-29 | 2017-12-29 | 腾讯科技(深圳)有限公司 | The recommendation method and apparatus of application program |
CN107644079A (en) * | 2015-05-22 | 2018-01-30 | 广东欧珀移动通信有限公司 | Method and device and related media production are recommended in one kind application |
CN107644042A (en) * | 2016-07-22 | 2018-01-30 | 平安科技(深圳)有限公司 | Software program clicking rate estimates sort method and server |
CN107862022A (en) * | 2017-10-31 | 2018-03-30 | 中国科学院自动化研究所 | Cultural resource commending system |
CN108024148A (en) * | 2016-10-31 | 2018-05-11 | 腾讯科技(深圳)有限公司 | The multimedia file recognition methods of Behavior-based control feature, processing method and processing device |
CN105227445B (en) * | 2015-10-23 | 2018-09-14 | 中国联合网络通信集团有限公司 | Recommend platform using methods and applications are recommended |
CN108833458A (en) * | 2018-04-02 | 2018-11-16 | 腾讯科技(深圳)有限公司 | A kind of application recommended method, device, medium and equipment |
CN109150983A (en) * | 2018-07-27 | 2019-01-04 | 腾讯科技(深圳)有限公司 | The control method and device of control method and device, data recommendation that front end is shown |
CN109151608A (en) * | 2018-09-06 | 2019-01-04 | 武汉斗鱼网络科技有限公司 | A kind of method and terminal of the sequence of face value list |
CN110152309A (en) * | 2019-01-24 | 2019-08-23 | 腾讯科技(深圳)有限公司 | Voice communication method, device, electronic equipment and storage medium |
CN110472071A (en) * | 2019-07-03 | 2019-11-19 | 中移(杭州)信息技术有限公司 | Multimedia file recommendation method, device, the network equipment and storage medium |
CN110827109A (en) * | 2018-08-14 | 2020-02-21 | 中国移动通信集团浙江有限公司 | Short message marketing method and device |
CN114979083A (en) * | 2022-05-13 | 2022-08-30 | 杭州网易云音乐科技有限公司 | Audio playing method, audio playing device, equipment, medium and computing equipment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6266649B1 (en) * | 1998-09-18 | 2001-07-24 | Amazon.Com, Inc. | Collaborative recommendations using item-to-item similarity mappings |
CN102567511A (en) * | 2011-12-27 | 2012-07-11 | 奇智软件(北京)有限公司 | Method and device for automatically recommending application |
CN102685565A (en) * | 2012-05-18 | 2012-09-19 | 合一网络技术(北京)有限公司 | Click feedback type individual recommendation system |
-
2012
- 2012-11-15 CN CN2012104620790A patent/CN102999588A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6266649B1 (en) * | 1998-09-18 | 2001-07-24 | Amazon.Com, Inc. | Collaborative recommendations using item-to-item similarity mappings |
CN102567511A (en) * | 2011-12-27 | 2012-07-11 | 奇智软件(北京)有限公司 | Method and device for automatically recommending application |
CN102685565A (en) * | 2012-05-18 | 2012-09-19 | 合一网络技术(北京)有限公司 | Click feedback type individual recommendation system |
Non-Patent Citations (1)
Title |
---|
胡迎松等: "一个基于Agent的个性化推荐系统", 《计算机应用研究》 * |
Cited By (63)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104182449A (en) * | 2013-05-20 | 2014-12-03 | Tcl集团股份有限公司 | System and method for personalized video recommendation based on user interests modeling |
CN104182449B (en) * | 2013-05-20 | 2018-10-26 | Tcl集团股份有限公司 | Individualized video commending system and method based on user interest modeling |
CN103297853A (en) * | 2013-06-07 | 2013-09-11 | 华东师范大学 | IPTV (internet protocol television) program recommendation method based on context recognition for multiple users |
CN103297853B (en) * | 2013-06-07 | 2016-04-06 | 华东师范大学 | A kind of IPTV program commending method based on multi-user's context identification |
CN103309984A (en) * | 2013-06-17 | 2013-09-18 | 腾讯科技(深圳)有限公司 | Data processing method and device |
CN103309984B (en) * | 2013-06-17 | 2016-12-28 | 腾讯科技(深圳)有限公司 | The method and apparatus that data process |
CN103593434A (en) * | 2013-11-12 | 2014-02-19 | 北京奇虎科技有限公司 | Application recommendation method and device and server equipment |
CN104636406B (en) * | 2013-11-15 | 2018-05-18 | 华为技术有限公司 | A kind of method and apparatus according to user behavior pushed information |
CN104636406A (en) * | 2013-11-15 | 2015-05-20 | 华为技术有限公司 | Method and device for pushing information according to user behaviors |
CN103678518B (en) * | 2013-11-28 | 2017-02-15 | 北京邮电大学 | Method and device for adjusting recommendation lists |
CN103685498B (en) * | 2013-12-06 | 2018-01-19 | 北京奇虎科技有限公司 | The loading method and browser device of a kind of wallpaper |
CN103685498A (en) * | 2013-12-06 | 2014-03-26 | 北京奇虎科技有限公司 | Wallpaper loading method and browser device |
CN105580385B (en) * | 2013-12-31 | 2018-12-25 | 广州华多网络科技有限公司 | Channel cut-in method and system |
CN105580385A (en) * | 2013-12-31 | 2016-05-11 | 广州华多网络科技有限公司 | Channel access method and system |
CN103702226A (en) * | 2013-12-31 | 2014-04-02 | 广州华多网络科技有限公司 | Displaying method and device of direct broadcasting client end channel information |
WO2015131803A1 (en) * | 2014-03-06 | 2015-09-11 | Tencent Technology (Shenzhen) Company Limited | Application recommending method and system |
CN104915354A (en) * | 2014-03-12 | 2015-09-16 | 深圳市腾讯计算机系统有限公司 | Multimedia file pushing method and device |
CN104915354B (en) * | 2014-03-12 | 2020-01-10 | 深圳市腾讯计算机系统有限公司 | Multimedia file pushing method and device |
US10108675B2 (en) | 2014-03-19 | 2018-10-23 | Huawei Technologies Co., Ltd. | Application recommending method and system, and server |
WO2015139232A1 (en) * | 2014-03-19 | 2015-09-24 | 华为技术有限公司 | Application recommendation method, system and server |
CN104603753A (en) * | 2014-03-19 | 2015-05-06 | 华为技术有限公司 | Method, system and server for recommending application |
US10956424B2 (en) | 2014-03-19 | 2021-03-23 | Huawei Technologies Co., Ltd. | Application recommending method and system, and server |
CN103955484A (en) * | 2014-04-09 | 2014-07-30 | 微梦创科网络科技(中国)有限公司 | Method and system for estimating tendency of user to network social tools |
CN103955484B (en) * | 2014-04-09 | 2017-06-06 | 微梦创科网络科技(中国)有限公司 | Tendentious appraisal procedure and system of the user to network social intercourse instrument |
CN105095305A (en) * | 2014-05-20 | 2015-11-25 | 深圳市腾讯计算机系统有限公司 | Method and device for generating personalized page |
CN105095305B (en) * | 2014-05-20 | 2019-03-08 | 深圳市腾讯计算机系统有限公司 | A kind of generation method and device of personal page |
CN104065981A (en) * | 2014-06-20 | 2014-09-24 | 海信集团有限公司 | Method and device for recommending videos |
CN104243590A (en) * | 2014-09-19 | 2014-12-24 | 广州华多网络科技有限公司 | Resource object recommendation method and device |
CN104243590B (en) * | 2014-09-19 | 2017-10-13 | 广州华多网络科技有限公司 | Resource object recommends method and apparatus |
CN105740268A (en) * | 2014-12-10 | 2016-07-06 | 阿里巴巴集团控股有限公司 | Information pushing method and apparatus |
CN105740268B (en) * | 2014-12-10 | 2019-04-09 | 阿里巴巴集团控股有限公司 | A kind of information-pushing method and device |
CN104735150A (en) * | 2015-03-27 | 2015-06-24 | 努比亚技术有限公司 | Message pushing method and device |
CN107644079A (en) * | 2015-05-22 | 2018-01-30 | 广东欧珀移动通信有限公司 | Method and device and related media production are recommended in one kind application |
CN105227445B (en) * | 2015-10-23 | 2018-09-14 | 中国联合网络通信集团有限公司 | Recommend platform using methods and applications are recommended |
CN105307023A (en) * | 2015-10-30 | 2016-02-03 | 北京奇艺世纪科技有限公司 | Application recommending method and apparatus, and video recommending method and apparatus |
CN105307023B (en) * | 2015-10-30 | 2018-04-13 | 北京奇艺世纪科技有限公司 | Using recommendation method, apparatus and video recommendation method, device |
CN105868248A (en) * | 2015-12-15 | 2016-08-17 | 乐视网信息技术(北京)股份有限公司 | Media recommendation method and device |
CN106951418A (en) * | 2016-01-07 | 2017-07-14 | 广州启法信息科技有限公司 | A kind of personalized recommendation technology of business web site |
CN107045688A (en) * | 2016-02-05 | 2017-08-15 | 上海行邑信息科技有限公司 | A kind of control method and device of accurate identification target terminal |
CN105787053A (en) * | 2016-02-26 | 2016-07-20 | 维沃移动通信有限公司 | Application pushing method and terminal |
CN105930423A (en) * | 2016-04-18 | 2016-09-07 | 乐视控股(北京)有限公司 | Multimedia similarity determination method and apparatus as well as multimedia recommendation method |
CN107306355A (en) * | 2016-04-20 | 2017-10-31 | 华为技术有限公司 | A kind of content recommendation method and server |
CN107306355B (en) * | 2016-04-20 | 2019-10-25 | 华为技术有限公司 | A kind of content recommendation method and server |
CN107644042B (en) * | 2016-07-22 | 2020-06-30 | 平安科技(深圳)有限公司 | Software program click rate pre-estimation sorting method and server |
CN107644042A (en) * | 2016-07-22 | 2018-01-30 | 平安科技(深圳)有限公司 | Software program clicking rate estimates sort method and server |
CN107526753B (en) * | 2016-07-29 | 2020-06-16 | 腾讯科技(深圳)有限公司 | Recommendation method and device for application program |
CN107526753A (en) * | 2016-07-29 | 2017-12-29 | 腾讯科技(深圳)有限公司 | The recommendation method and apparatus of application program |
CN106484766B (en) * | 2016-09-07 | 2019-10-22 | 北京百度网讯科技有限公司 | Searching method and device based on artificial intelligence |
CN106484766A (en) * | 2016-09-07 | 2017-03-08 | 北京百度网讯科技有限公司 | Searching method based on artificial intelligence and device |
CN108024148B (en) * | 2016-10-31 | 2020-02-28 | 腾讯科技(深圳)有限公司 | Behavior feature-based multimedia file identification method, processing method and device |
CN108024148A (en) * | 2016-10-31 | 2018-05-11 | 腾讯科技(深圳)有限公司 | The multimedia file recognition methods of Behavior-based control feature, processing method and processing device |
CN107330747A (en) * | 2017-05-16 | 2017-11-07 | 深圳和而泰智能家居科技有限公司 | Beauty appliance gear recommends method, beauty appliance and storage medium |
CN107862022B (en) * | 2017-10-31 | 2020-12-11 | 中国科学院自动化研究所 | Culture resource recommendation system |
CN107862022A (en) * | 2017-10-31 | 2018-03-30 | 中国科学院自动化研究所 | Cultural resource commending system |
CN108833458A (en) * | 2018-04-02 | 2018-11-16 | 腾讯科技(深圳)有限公司 | A kind of application recommended method, device, medium and equipment |
CN109150983A (en) * | 2018-07-27 | 2019-01-04 | 腾讯科技(深圳)有限公司 | The control method and device of control method and device, data recommendation that front end is shown |
CN110827109B (en) * | 2018-08-14 | 2022-07-29 | 中国移动通信集团浙江有限公司 | Short message marketing method and device |
CN110827109A (en) * | 2018-08-14 | 2020-02-21 | 中国移动通信集团浙江有限公司 | Short message marketing method and device |
CN109151608A (en) * | 2018-09-06 | 2019-01-04 | 武汉斗鱼网络科技有限公司 | A kind of method and terminal of the sequence of face value list |
CN110152309B (en) * | 2019-01-24 | 2021-10-26 | 腾讯科技(深圳)有限公司 | Voice communication method, device, electronic equipment and storage medium |
CN110152309A (en) * | 2019-01-24 | 2019-08-23 | 腾讯科技(深圳)有限公司 | Voice communication method, device, electronic equipment and storage medium |
CN110472071A (en) * | 2019-07-03 | 2019-11-19 | 中移(杭州)信息技术有限公司 | Multimedia file recommendation method, device, the network equipment and storage medium |
CN114979083A (en) * | 2022-05-13 | 2022-08-30 | 杭州网易云音乐科技有限公司 | Audio playing method, audio playing device, equipment, medium and computing equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102999588A (en) | Method and system for recommending multimedia applications | |
CN107004205B (en) | System and method for suggesting creative types for online content items to advertisers | |
CN105791157B (en) | A kind of distribution method of flow, distribution system and server | |
CN108595461B (en) | Interest exploration method, storage medium, electronic device and system | |
CN103024017B (en) | A kind of social networks important goal and Community Group recognition methods | |
CN108140041B (en) | For the viewing time cluster of video search | |
WO2017071251A1 (en) | Information pushing method and device | |
CN105005582B (en) | The recommendation method and device of multimedia messages | |
CN105898209A (en) | Video platform monitoring and analyzing system | |
US11379540B2 (en) | Gain adjustment component for computer network routing infrastructure | |
US9280749B1 (en) | Determining an attribute of an online user using user device data | |
CN101482884A (en) | Cooperation recommending system based on user predilection grade distribution | |
CN103106208A (en) | Streaming media content recommendation method and system in mobile internet | |
CN103593350A (en) | Method and device for recommending promotion keyword price parameters | |
CN103309894B (en) | Based on search implementation method and the system of user property | |
CN106326297B (en) | Application program recommendation method and device | |
CN112446727B (en) | Advertisement triggering method, device, equipment and computer readable storage medium | |
CN108965360B (en) | Recommended data processing method, server and computer storage medium | |
CN111523035A (en) | Recommendation method, device, server and medium for APP browsing content | |
CN106257449B (en) | A kind of information determines method and apparatus | |
CN104102727A (en) | Query term recommending method and device | |
CN106484701B (en) | A kind of information displaying method and device based on business object | |
Li et al. | Multimedia cloud content distribution based on interest discovery and integrated utility of user | |
CN113014938B (en) | Multi-dimensional live video recommendation method and device | |
CN113034231B (en) | Multi-supply chain commodity intelligent recommendation system and method based on SaaS cloud service |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C12 | Rejection of a patent application after its publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130327 |