CN102651033B - A kind of recommend method of online resource and device - Google Patents
A kind of recommend method of online resource and device Download PDFInfo
- Publication number
- CN102651033B CN102651033B CN201210101954.2A CN201210101954A CN102651033B CN 102651033 B CN102651033 B CN 102651033B CN 201210101954 A CN201210101954 A CN 201210101954A CN 102651033 B CN102651033 B CN 102651033B
- Authority
- CN
- China
- Prior art keywords
- user
- resource
- quality
- online
- success ratio
- 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.)
- Active
Links
Abstract
The invention provides a kind of recommend method and device of online resource, based on Quality Control Model, described Quality Control Model comprises: utilize the qualitative data of user's history to carry out user quality classification submodel that cluster obtains, and the online use success ratio of each user quality that each resource class utilizing the resource usage data of user's history to set up is corresponding calculates submodel; Described recommend method comprises: S1, real-time quality data when sending request according to active user, utilize user quality to classify quality category that submodel determines belonging to active user; S2, utilize the online success ratio that uses to calculate submodel, determine the online use success ratio of each resource quality that quality category belonging to active user is corresponding; S3, the online use success ratio of each resource quality determined according to step S2, recommend the online resource of asking to described active user.The use success ratio of online resource can be improved by the present invention, bring good experience to user.
Description
[technical field]
The present invention relates to technical field of the computer network, particularly a kind of recommend method of online resource and device.
[background technology]
Along with the lifting of the network bandwidth and the change of user habit, now the demand of Internet resources is mainly utilized online, such as online audio plays, online broadcasting video etc.When user needs to obtain certain online resource, the mode of usual employing vertical search, resource information is inputted in the search box of browser, search server is sent in the request that this resource information comprises by browser, return the online resource corresponding with this resource information by search server to browser, and recommend user for user and select online utilization.Such as, user inputs " king of K song " and clicks audio types to carry out vertical search in the search box of browser, after request is sent to search server by browser, just can get the online song of " king of K song " correspondence that search server returns, and recommend user for user's online test listening.
But, in prior art when online resource is recommended user, normally based on online resource quality, the readability of such as online resource, the downloading rate etc. of online resource.This will cause, although it has higher quality although come online resource above, because user quality is lower, the network speed of such as user is lower, and cannot the high-quality online resource of smooth playing, makes the use failure of online resource.Visible, the use success ratio that the existing online resource way of recommendation is brought is lower, and Consumer's Experience is poor.
[summary of the invention]
In view of this, the invention provides a kind of recommend method and device of online resource, so that improve the use success ratio of online resource, thus bring good experience to user.
Concrete technical scheme is as follows:
A kind of recommend method of online resource, based on Quality Control Model, described Quality Control Model comprises: utilize the qualitative data of user's history to carry out user quality classification submodel that cluster obtains, and the online use success ratio of each user quality that each resource class utilizing the resource usage data of user's history to set up is corresponding calculates submodel; Described recommend method comprises:
S1, real-time quality data when sending request according to active user, utilize described user quality to classify quality category that submodel determines belonging to active user;
S2, utilize described online use success ratio to calculate submodel, determine the online use success ratio of each resource quality that quality category belonging to described active user is corresponding;
S3, the online use success ratio of each resource quality determined according to described step S2, recommend the online resource of asking to described active user.
According to one preferred embodiment of the present invention, described qualitative data comprises at least one in user side network speed, request time or positional information.
According to one preferred embodiment of the present invention, the method also comprises:
Use user in the process of online resource and collect qualitative data, and after use terminates, the qualitative data collected is carried out recording or reporting server with the form of daily record, for setting up described user quality classification submodel.
According to one preferred embodiment of the present invention, set up described online use success ratio calculating submodel specifically to comprise: utilize the online resource of user's history to obtain the data of success or not, statistics different quality class users is respectively for the use success ratio of each resource class.
According to one preferred embodiment of the present invention, described step S3 specifically comprises: from the online resource that described active user asks, and selects to meet the online resource presetting the online resource quality using success ratio to require and recommends to described active user; Or,
Carry out the backward described active user of sequence according to the online online resource using success ratio described active user to be asked to recommend; Or,
Meeting under the prerequisite presetting the requirement of online use success ratio, recommend the online resource of asking to described active user according to online resource quality height.
A kind of recommendation apparatus of online resource, based on Quality Control Model, described Quality Control Model comprises: utilize the qualitative data of user's history to carry out user quality classification submodel that cluster obtains, and the online use success ratio of each user quality that each resource class utilizing the resource usage data of user's history to set up is corresponding calculates submodel; Described recommendation apparatus comprises:
Data collection module, for collecting the real-time quality data of user;
Quality category determining unit, real-time quality data during for sending request according to active user, utilize described user quality to classify quality category that submodel determines belonging to active user;
Success ratio determining unit, for utilizing described online use success ratio to calculate submodel, determines the online use success ratio of each resource quality that quality category belonging to described active user is corresponding;
Resource recommendation unit, for the online use success ratio of each resource quality determined according to described success ratio determining unit, recommends the online resource of asking to described active user.
According to one preferred embodiment of the present invention, described qualitative data comprises at least one in user side network speed, request time or positional information.
According to one preferred embodiment of the present invention, this recommendation apparatus also comprises: unit set up by the first model, carries out cluster obtain user quality classification submodel for utilizing the qualitative data of user's history;
Described data collection module, also collect qualitative data for using in the process of online resource user, and after use terminates, the qualitative data collected is carried out recording or reporting server with the form of daily record, set up unit for described first model and obtain described user quality classification submodel.
According to one preferred embodiment of the present invention, this recommendation apparatus also comprises: unit set up by the second model, for the data utilizing the online resource of user's history to obtain success or not, statistics different quality class users respectively for the use success ratio of each resource class, thus sets up online use success ratio calculating submodel.
According to one preferred embodiment of the present invention, described resource recommendation unit, from the online resource that described active user asks, is selected to meet the online resource presetting the online resource quality using success ratio to require and is recommended to described active user; Or,
Carry out the backward described active user of sequence according to the online online resource using success ratio described active user to be asked to recommend; Or,
Meeting under the prerequisite presetting the requirement of online use success ratio, recommend the online resource of asking to described active user according to online resource quality height.
As can be seen from the above technical solutions, real-time quality data when first recommend method provided by the invention and device send request according to active user determine the quality category belonging to user, determine when mating with the quality category of this user again from the online resource that active user asks, meet the online resource that online success ratio requires and recommend.Thus ensure that online resource successfully uses, and namely improves use success ratio and the Consumer's Experience of online resource as much as possible.
[accompanying drawing explanation]
The recommend method process flow diagram of the online resource that Fig. 1 provides for the embodiment of the present invention;
The recommendation apparatus structural drawing of the online resource that Fig. 2 provides for the embodiment of the present invention.
[embodiment]
In order to make the object, technical solutions and advantages of the present invention clearly, describe the present invention below in conjunction with the drawings and specific embodiments.
Find by analyzing, often user is when utilizing online resource, major demands plays success ratio and fluency online, next is only other demands such as resource quality (such as tonequality, sharpness), if therefore will success ratio be improved, then need to realize mating of resource quality and user quality, this fits through Quality Control Model and judges.The present invention proposes based on this, is first described two Seed models involved by Quality Control Model: user quality classification submodel and the online success ratio that uses calculate submodel.
User quality classification submodel utilizes the qualitative data of user's history to carry out cluster to obtain, and wherein the qualitative data of user's history obtains from online resource usage log.The real-time quality data that user asks online resource at every turn can be collected, include but not limited to: at least one in the network speed (user uses the network speed of online resource) of user side, request time or positional information, wherein each user can adopt cookie or user name etc. to identify, request time can be date and time information, week information or period information etc., and positional information can be ip address or geographical location information etc.The mode of collecting these qualitative datas by embedding js code in web player, can use user in the process of online resource and collecting qualitative data, and after use terminates, the information collected being carried out recording or reporting server with the form of daily record.
When carrying out cluster, carry out according to the qualitative data collected, same user can be divided into more than one and plant quality category.Such as:
According to the network speed of each user when using online resource, user is divided into highspeed user, middling speed user and low speed user, wherein network speed is that the user of more than 60kbps is divided into highspeed user, network speed is that the user of 30kbps-60kbps is divided into middling speed user, and network speed is that the user of 0-30kbps is divided into low speed user.
According to the request time of each user to online resource, user is divided into user's peak period, ordinary period user and unexpected winner period user.Wherein, request time is at 10 and is divided into user's peak period at 17 in afternoon and at 8 in evening to the evening user of 10, request time is that the user of 2:00 AM to 8 is divided into unexpected winner period user, and request time is that the user of other periods is divided into ordinary period user.Also directly user can be divided according to the period at request time place, such as be divided into respectively: 0 o'clock user to 3 o'clock period, 4 o'clock users to 7 o'clock period, 8 o'clock users to 11 o'clock period, 12 o'clock users to 15 o'clock period, 16 o'clock users to 19 o'clock period, 20 o'clock users to 23 o'clock period.
According to the positional information of each user, user is divided into the user of each province.
Also can carry out integration to the request time of user, user side network speed and positional information and be divided into large class, such as, network speed is more than 60kbps, request time is divided into a class 10 Beijing users at 17 in afternoon.Certain class of subscriber can also adopt other to divide the mode of quality category, illustrates no longer one by one at this.In addition, can adopt the mode of above-mentioned absolute classification, also can adopt and calculate based on real-time quality data the mode corresponding to the probability of each quality category, the concrete mode classification embodiment of the present invention of user quality classification is not limited.
It is utilize the resource usage data of user's history to set up that online use success ratio calculates submodel, after the classification each user being carried out to quality category, the online resource of user's history is utilized to obtain the data of success or not, statistics different quality class users respectively for the use success ratio of each resource class, the ratio of the b class resource quantity that a class user can adopt a class user to use the successful online resource quantity of b class resource and a class user to ask for the use success ratio of b class resource.
Wherein the resource class of online resource can divide according to the quality of online resource, such as, according to the size or service speed etc. of resource, each online resource is divided into high-quality resource, fair average quality resource and inferior quality resource.Wherein the service speed of resource is usually by the embodiment of testing the speed of resource place content distributing network (CDN, ContentDeliveryNetwork) node server.
Such as, obtain that network speed is more than 60kbps through statistics, request time is 60% 10 Beijing users at 17 in afternoon for the use success ratio of high-quality resource, is 92% for the use success ratio of fair average quality resource, is 98% for the use success ratio of inferior quality resource.
Be described the method how utilizing above-mentioned submodel to realize online resource recommendation below in conjunction with Fig. 1, as shown in Figure 1, the method comprises the following steps:
Step 101: real-time quality data when asking online resource according to active user, utilize user quality to classify quality category that submodel determines belonging to active user.
Active user asks real-time quality data during online resource to include but not limited to: at least one in user side network speed (real-time network speed), request time or positional information, after real-time quality data input user quality classification submodel when active user being asked online resource, the quality category belonging to active user can be obtained.
Such as, active user asks real-time quality data during online resource to be: network speed 70kbps in real time, request time is 13 points, positional information is Beijing, then after inputting user quality classification submodel, active user being divided into real-time network speed is that more than 60kbps, request time are 10 these classifications of Beijing user at 17 in afternoon.
Step 102: utilize the online success ratio that uses to calculate submodel, determines the online use success ratio of each resource quality that quality category belonging to active user is corresponding.
Continue upper example, be that more than 60kbps, request time are at this classification Input Online of 10 Beijing users at 17 in afternoon use success ratio calculating submodel by real-time network speed, the online use success ratio of each resource quality that quality category belonging to active user is corresponding can be determined, the use success ratio namely for high-quality resource is 60%, be 92% for the use success ratio of fair average quality resource, be 98% for the use success ratio of inferior quality resource.
Step 103: according to the online use success ratio of each resource quality that step 102 is determined, recommend the online resource of asking to active user.
In this step, first the online resource that user asks can be determined, namely the online resource of all correspondences is determined according to the resource information of user's input, the resource quality of these online resources differs, such as user's input " king of K song " when clicking audio types to carry out vertical search, first determine the audio resource that all " king of K song " hits, but high-quality resource, fair average quality resource and inferior quality resource in these resources, may be there is.So how these online resources are recommended to user, then can adopt different strategies, include but not limited to following several:
The online resource that Generalization bounds one, selection meet the resource quality presetting the requirement of online use success ratio is recommended to active user.
Suppose presetting online use success ratio requires to be more than 90%, the real-time network speed then mentioned in upper example is that more than 60kbps, request time are 10 Beijing users at 17 in afternoon, fair average quality resource and inferior quality resource recommendation can be selected to active user, and the use success ratio of high-quality resource is owing to presetting online below the success ratio requirement that uses, therefore active user can not recommended.
Generalization bounds two, the backward active user that carries out sorting according to the online online resource using success ratio user to be asked recommend.That is, the online resource that all users ask all is recommended user, but according to the online success ratio that uses, online resource is sorted when recommending.
Generalization bounds three, under meeting and presetting the online prerequisite using success ratio to require, recommend the online resource of asking to active user according to online resource quality height.
Still suppose presetting online use success ratio requires to be more than 90%, the real-time network speed then mentioned in upper example is that more than 60kbps, request time are 10 Beijing users at 17 in afternoon, fair average quality resource and inferior quality resource recommendation can be selected to active user, for active user's prioritizing selection before fair average quality resource being come low quality resource such as grade in recommendation process.
It is more than the detailed description that method provided by the present invention is carried out, below device provided by the present invention is described, equally, the recommendation apparatus of online resource provided by the present invention is based on Quality Control Model, specifically comprise: utilize the qualitative data of user's history to carry out user quality classification submodel that cluster obtains, and the online use success ratio of each user quality that each resource class utilizing the resource usage data of user's history to set up is corresponding calculates submodel.As shown in Figure 2, the recommendation apparatus of online resource can comprise: data collection module 201, quality category determining unit 202, success ratio determining unit 203 and resource recommendation unit 204.
Data collection module 201 collects the qualitative data of user.Wherein, qualitative data can comprise at least one in user side network speed, request time or positional information.Each user can adopt cookie or user name etc. to identify.Collect these qualitative datas and can deliver to server by the js code execution concurrence embedded in web player.
Real-time quality data when quality category determining unit 202 sends request according to active user, utilize user quality to classify quality category that submodel determines belonging to active user.Now real-time quality data comprise at least one in the real-time network speed of user side, request time or positional information.
Success ratio determining unit 203 utilizes the online success ratio that uses to calculate submodel, determines the online use success ratio of each resource quality that quality category belonging to active user is corresponding.
The online use success ratio of each resource quality that resource recommendation unit 204 is determined according to success ratio determining unit 203, recommends the online resource of asking to active user.First the online resource that user asks can be determined, namely the online resource of all correspondences is determined according to the resource information of user's input, the resource quality of these online resources differs, and when being recommended to user by these online resources, can adopt but be not limited to following several strategy:
Generalization bounds one, from the online resource that active user asks, select to meet the online resource presetting the online resource quality using success ratio to require and recommend to active user.
Generalization bounds two, the backward active user that carries out sorting according to the online online resource using success ratio active user to be asked recommend.
Generalization bounds three, under meeting and presetting the online prerequisite using success ratio to require, recommend the online resource of asking to active user according to online resource quality height.
In order to realize the foundation of above-mentioned two Seed models, this recommendation apparatus can further include: unit 205 set up by the first model, carries out cluster obtain user quality classification submodel for utilizing the qualitative data of user's history.When carrying out cluster, carry out according to the qualitative data collected, same user can be divided into more than one and plant quality category.
Data collection module 201, also collect qualitative data for using in the process of online resource user, and after use terminates, the qualitative data collected is carried out recording or reporting server with the form of daily record, set up unit 205 for the first model and obtain user quality classification submodel.That is, when each user uses online resource, realize while online resource recommends utilizing real-time qualitative data, the qualitative data of collection also as daily record for the renewal of subsequent user quality classification submodel, thus achieve a closed-loop system.
In addition, this recommendation apparatus also comprises: unit 206 set up by the second model, for the data utilizing the online resource of user's history to obtain success or not, statistics different quality class users is respectively for the use success ratio of each resource class, thus the online success ratio that uses of foundation calculates submodel.Wherein, the data of the online resource acquisition success or not of user's history also can be obtained by data collection module 201.
Above-mentioned recommend method provided by the invention and recommendation apparatus realize at server end, after receiving the request that user sent by browser or client, to the online resource that browser or client recommend user to ask.
In addition, it should be noted that, the online resource that the present invention relates to can include but not limited to: online audio resource, Online Video resource etc.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment made, equivalent replacement, improvement etc., all should be included within the scope of protection of the invention.
Claims (10)
1. the recommend method of an online resource, based on Quality Control Model, it is characterized in that, described Quality Control Model comprises: utilize the qualitative data of user's history to carry out user quality classification submodel that cluster obtains, and the online use success ratio of each user quality that each resource class utilizing the resource usage data of user's history to set up is corresponding calculates submodel; Described recommend method comprises:
S1, real-time quality data when sending request according to active user, utilize described user quality to classify quality category that submodel determines belonging to active user;
S2, utilize described online use success ratio to calculate submodel, determine the online use success ratio of each resource quality that quality category belonging to described active user is corresponding;
S3, the online use success ratio of each resource quality determined according to described step S2, recommend the online resource of asking to described active user.
2. method according to claim 1, is characterized in that, described qualitative data comprises at least one in user side network speed, request time or positional information.
3. method according to claim 1 and 2, is characterized in that, the method also comprises:
Use user in the process of online resource and collect qualitative data, and after use terminates, the qualitative data collected is carried out recording or reporting server with the form of daily record, for setting up described user quality classification submodel.
4. method according to claim 1, it is characterized in that, set up described online use success ratio calculating submodel specifically to comprise: utilize the online resource of user's history to obtain the data of success or not, statistics different quality class users is respectively for the use success ratio of each resource class.
5. method according to claim 1, it is characterized in that, described step S3 specifically comprises: from the online resource that described active user asks, and selects to meet the online resource presetting the online resource quality using success ratio to require and recommends to described active user; Or,
Carry out the backward described active user of sequence according to the online online resource using success ratio described active user to be asked to recommend; Or,
Meeting under the prerequisite presetting the requirement of online use success ratio, recommend the online resource of asking to described active user according to online resource quality height.
6. the recommendation apparatus of an online resource, based on Quality Control Model, it is characterized in that, described Quality Control Model comprises: utilize the qualitative data of user's history to carry out user quality classification submodel that cluster obtains, and the online use success ratio of each user quality that each resource class utilizing the resource usage data of user's history to set up is corresponding calculates submodel; Described recommendation apparatus comprises:
Data collection module, for collecting the qualitative data of user;
Quality category determining unit, real-time quality data during for sending request according to active user, utilize described user quality to classify quality category that submodel determines belonging to active user;
Success ratio determining unit, for utilizing described online use success ratio to calculate submodel, determines the online use success ratio of each resource quality that quality category belonging to described active user is corresponding;
Resource recommendation unit, for the online use success ratio of each resource quality determined according to described success ratio determining unit, recommends the online resource of asking to described active user.
7. recommendation apparatus according to claim 6, is characterized in that, described qualitative data comprises at least one in user side network speed, request time or positional information.
8. the recommendation apparatus according to claim 6 or 7, is characterized in that, this recommendation apparatus also comprises: unit set up by the first model, carries out cluster obtain user quality classification submodel for utilizing the qualitative data of user's history;
Described data collection module, also collect qualitative data for using in the process of online resource user, and after use terminates, the qualitative data collected is carried out recording or reporting server with the form of daily record, set up unit for described first model and obtain described user quality classification submodel.
9. recommendation apparatus according to claim 6, it is characterized in that, this recommendation apparatus also comprises: unit set up by the second model, for the data utilizing the online resource of user's history to obtain success or not, statistics different quality class users respectively for the use success ratio of each resource class, thus sets up online use success ratio calculating submodel.
10. recommendation apparatus according to claim 6, it is characterized in that, described resource recommendation unit, from the online resource that described active user asks, is selected to meet the online resource presetting the online resource quality using success ratio to require and is recommended to described active user; Or,
Carry out the backward described active user of sequence according to the online online resource using success ratio described active user to be asked to recommend; Or,
Meeting under the prerequisite presetting the requirement of online use success ratio, recommend the online resource of asking to described active user according to online resource quality height.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210101954.2A CN102651033B (en) | 2012-04-09 | 2012-04-09 | A kind of recommend method of online resource and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210101954.2A CN102651033B (en) | 2012-04-09 | 2012-04-09 | A kind of recommend method of online resource and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102651033A CN102651033A (en) | 2012-08-29 |
CN102651033B true CN102651033B (en) | 2016-04-27 |
Family
ID=46693041
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210101954.2A Active CN102651033B (en) | 2012-04-09 | 2012-04-09 | A kind of recommend method of online resource and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102651033B (en) |
Families Citing this family (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103685397B (en) * | 2012-09-14 | 2017-04-05 | 百度在线网络技术(北京)有限公司 | Based on the reminding method of the download/play quality of high in the clouds analysis, system and device |
TWI538491B (en) * | 2013-09-02 | 2016-06-11 | 大同股份有限公司 | Television service system and method for supplying video service |
EP3164842A4 (en) * | 2014-07-01 | 2017-11-08 | Google LLC | Systems and methods for suggesting creative types for online content items to advertiser |
CN104363302A (en) * | 2014-11-27 | 2015-02-18 | 北京奇艺世纪科技有限公司 | Method and device for downloading file offline |
CN104768069B (en) * | 2015-04-02 | 2017-10-03 | 天脉聚源(北京)传媒科技有限公司 | A kind of data projection method, apparatus and system |
CN105160225B (en) * | 2015-08-18 | 2018-07-27 | 宇龙计算机通信科技(深圳)有限公司 | Electronic device method and device |
CN108415992B (en) * | 2018-02-12 | 2022-03-04 | 百度在线网络技术(北京)有限公司 | Resource recommendation method and device and computer equipment |
CN108550046B (en) * | 2018-03-07 | 2021-03-02 | 创新先进技术有限公司 | Resource and marketing recommendation method and device and electronic equipment |
CN108647293B (en) * | 2018-05-07 | 2022-02-01 | 广州虎牙信息科技有限公司 | Video recommendation method and device, storage medium and server |
CN111385659B (en) * | 2018-12-29 | 2021-08-17 | 广州市百果园信息技术有限公司 | Video recommendation method, device, equipment and storage medium |
CN111311357A (en) * | 2020-01-20 | 2020-06-19 | 昊居科技有限公司 | House transaction information management method and system |
CN112559873B (en) * | 2020-12-21 | 2021-08-13 | 融易学控股(深圳)有限公司 | User recommendation system based on intelligent education |
CN115378878B (en) * | 2021-05-21 | 2023-11-14 | 北京字跳网络技术有限公司 | CDN scheduling method, device, equipment and storage medium |
CN114092162B (en) * | 2022-01-21 | 2022-07-01 | 北京达佳互联信息技术有限公司 | Recommendation quality determination method, and training method and device of recommendation quality determination model |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW200923807A (en) * | 2007-11-23 | 2009-06-01 | Inst Information Industry | Method and system for searching knowledge owner in network community |
CN101615197B (en) * | 2009-07-30 | 2011-08-31 | 浙江大学 | Personalized network resource recommended method based on network connection speed |
CN102054003B (en) * | 2009-11-04 | 2014-03-19 | 北京搜狗科技发展有限公司 | Methods and systems for recommending network information and creating network resource index |
-
2012
- 2012-04-09 CN CN201210101954.2A patent/CN102651033B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN102651033A (en) | 2012-08-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102651033B (en) | A kind of recommend method of online resource and device | |
JP6612949B2 (en) | Method, apparatus and storage medium for sharing online media impression data | |
US20170142482A1 (en) | Video platform monitoring and analyzing system | |
CN110222880B (en) | Service risk determining method, model training method and data processing method | |
CN101247402B (en) | Multimedia files downloading and broadcasting system and method | |
CN1666201B (en) | Method for the real-time distribution of streaming data on a network | |
CN110363621B (en) | Order information pushing system based on artificial intelligence technology | |
CA2493190A1 (en) | Media data usage measurement and reporting systems and methods | |
WO2021051839A1 (en) | Data processing method, apparatus and system and storage medium | |
CN106028075A (en) | Bullet screen displaying method and device | |
CN108320168B (en) | Data analysis method and device | |
CN105163142A (en) | User preference determination method, video recommendation method, user preference determination system and video recommendation system | |
CN107679922A (en) | A kind of Internet advertising transaction delivery system and method | |
CN104618740A (en) | Multimedia order system based on Cloud computing environment | |
US20100100589A1 (en) | Apparatus and method for measuring advertising metrics | |
CN106027673A (en) | Resource pre-push control and training method and device | |
CN112199521B (en) | Automatic media asset arranging method, device and system, electronic device and storage medium | |
CN104410877A (en) | Method and device for processing user behavior data of network television | |
CN101276377B (en) | Method, system for acquiring resource related information and application in search engine | |
CN114268512B (en) | Chat room recommendation method, chat room recommendation device, chat room recommendation server, chat room recommendation terminal, chat room recommendation program and chat room recommendation storage medium | |
CN111143688B (en) | Evaluation method and system based on mobile news client | |
CN105229684A (en) | For controlling and optimize the system of the user-to-user information distribution in message exchange | |
CN110022484A (en) | A kind of methods of exhibiting, device, server and medium | |
CN115982449B (en) | Intelligent pushing optimization method based on platform big data feedback | |
JP6791567B2 (en) | Devices, methods and storage media for sharing online media impression data |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |