CN103186586A - Pushing method and system for data service - Google Patents
Pushing method and system for data service Download PDFInfo
- Publication number
- CN103186586A CN103186586A CN2011104520139A CN201110452013A CN103186586A CN 103186586 A CN103186586 A CN 103186586A CN 2011104520139 A CN2011104520139 A CN 2011104520139A CN 201110452013 A CN201110452013 A CN 201110452013A CN 103186586 A CN103186586 A CN 103186586A
- Authority
- CN
- China
- Prior art keywords
- user
- preference
- data
- preference data
- service content
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
The application provides a pushing method and system for data service. The method comprises the steps of determining initial service content preference data of a user according to basic information of the user; setting a discussion subject for the user according to the initial service content preference data of the user and inviting the user to join in the discussion; modifying the initial service content preference data of the user according to the correlation between the discussion records of the user and the initial service content preference data; and pushing the corresponding data service to the user according to the modified service content preference data. Since the pushing method provided by the application considers the discussion of the user, the success rate of pushing is improved.
Description
Technical field
The application relates to method for pushing and the system of data service.
Background technology
The propelling movement of data service can be divided into dual mode.A kind of mode is direct propelling movement mode, and this mode directly pushes various data services to the user by means such as note, outgoing calls.If the data service that user preference pushes then can be accepted this data service, otherwise will refuse.Be appreciated that this propelling movement mode owing to do not carry out propelling movement according to user's situation, the propelling movement amount is big, but the propelling movement success ratio is low.
Another kind of mode is to determine the data service of user institute preference according to user's essential information.But because only often also inconsistent with the data service of the preference of user's reality according to user's the determined user's of essential information preference, so the success ratio of this propelling movement mode is not high yet.
Summary of the invention
The application's purpose provides a kind of method for pushing and system that can partly improve a kind of data service of above-mentioned defective of the prior art at least, improves the success ratio that pushes.
According to the application's one side, a kind of method for pushing of data service is provided, comprising:
Determine user's initial service content preference data according to user basic information;
Initial service content preference data according to the user is set the discussion theme for the user, and invites the user to add discussion;
According to user's discussion record and the correlativity of initial service content preference data, user's initial service content preference data is revised; And
To push to the user with the corresponding data service of revising of business tine preference data.
According to the application on the other hand, provide a kind of supplying system of data service, having comprised:
Determination module is determined user's initial service content preference data according to user basic information;
Processing module, initial service content preference data according to the user is set the discussion theme for the user, invite the user to add discussion, according to user's discussion record and the correlativity of initial service content preference data, user's initial service content preference data is revised; And
Push module, will push to the user with the corresponding data service of revising of business tine preference data.
The application considers to know clearly user's discussion, thereby has improved the success ratio that data service pushes.
Description of drawings
Fig. 1 shows the process flow diagram according to the method for pushing of the data service of the application's embodiment.
Fig. 2 shows the method according to the foundation of the application's embodiment and training user preference network model.
Fig. 3 shows the exemplary analysis network according to the application's embodiment.
Fig. 4 shows the supplying system according to the data service of the application's embodiment.
Fig. 5 shows the threshold value setting module according to the application's embodiment.
Embodiment
In order to understand the application better, will make more detailed description to the application's various aspects with reference to the accompanying drawings.Be appreciated that described drawings and detailed description are the description to the application's preferred embodiment, but not limit the application's scope by any way.
Fig. 1 shows the process flow diagram according to the method for pushing of the data service of the application's embodiment.
As shown in Figure 1, the method for pushing 1000 of data service can may further comprise the steps.
According to a kind of embodiment, for realizing according to user's discussion record and the correlativity between the initial service content preference data, thereby the method that user's initial service content preference data is revised, can set the preference data correction threshold for the user, and with the number of times of institute's preference keyword of extracting and add up and set the preference data correction threshold and compare, and revise professional preference data according to the result of comparison.The number of times that the preference keyword that the statistics user occurs in record is discussed occurs, number of times and preference data correction threshold that the preference keyword is occurred compare to revise the initial service content preference data.The preference keyword refers to have the word of related service content-preference with representing the user.For example, the number of times that occurs as the preference keyword is lower than the preference data correction threshold, discussion record and the correlativity between the initial service content preference data that the user then is described are little, then deletion and the corresponding preference data of this preference keyword from the initial service content preference data; Otherwise, then keep or increase and the corresponding preference data of this preference keyword.
According to a kind of embodiment, the preference data correction threshold can be predefined.For example system can preestablish the preference data correction threshold according to user's information and to pushing the requirement of success ratio.
According to another kind of embodiment, can set up the user preference network model, and determine user's preference data correction threshold according to the network model of setting up.Can active user's business tine preference data be revised according to determined user's preference data correction threshold, and will push to the user with the corresponding data service of revised business tine preference data.Subsequently, the result of the success or not that can push according to the data service user preference network model that comes further training to set up.Be appreciated that this trained network model can comprehensively react user's situation, thereby improve the fiduciary level of determined user's preference data correction threshold, and further improve the success ratio that pushes.
Fig. 2 shows the method according to the foundation of the application's embodiment and training user preference network model.
In step 301, determine information and the propelling movement situation of known users.Table 1 has been represented known users 1,2..., and the situation of n, wherein, and preference 1, preference 2..., preference n represent 1,2...n kind according to revised business tine preference data and the data service that pushes to the user.Parameter " 1 " expression pushes successfully, and just refer to: the user has accepted and the corresponding data service of revised business tine preference data, i.e. the data service of Tui Songing is the data service of user's actual preferences, thereby this data service pushes successfully.
| |
Preference 2 | Preference 3 | Preference 4 | 。。 | Preference n | |
| The |
1 | 1 | 0 | 1 | 。。 | 0 |
| The user 2 | 1 | 0 | 0 | 1 | 。。 | ?1 |
| 。。。 | 。。。 | 。。。 | 。。。 | 。。。 | 。。。 | ?。。。 |
| User n | 0 | 1 | 0 | 1 | 。。 | ?1 |
Table 1
In step 302, according to determined user's information and propelling movement situation, set up and training user preference network model.For example, can set up exemplary analysis network as shown in Figure 3 according to existing user's situation.Wherein, the node in the network can comprise user basic information, business tine preference, preference threshold value etc.Limit between the node in the network, and the influence relation between the node is represented in the sensing of arrow.For example, user profile 1 is pointed to preference 1, and expression user profile 1 influences preference 1, and is the father node of preference 1.
For phase-split network, can and push the feature structure of definite networks such as successful situation according to existing user's master data, business tine preference.For example, can determine the feature structure of network by minimum description length (MDL) training algorithm.
The formula of minimum description length (MDL) training algorithm is as follows
Wherein Xi represents node, the father node set of Pa (Xi) expression Xi node, and N represents the bar number of user data; H and Entropy represent the information entropy that defines in the first agricultural theorem, can directly calculate according to agricultural formula earlier.
In specific embodiment, suppose to have two nodes of A, B in the predetermined phase-split network model, A represents to discuss the number of times that occurs with the diet correlation word in the record, and B represents to discuss the number of times that occurs with the clothes correlation word in the record.Then the network structure that can be made of two nodes of A, B has three kinds, that is: A and B are mutually independently, A points to B and B points to A.Suppose, for example, according to the discussion record that extracts the user, A=3, B=5 have been determined, value substitution formula 1 with A, B, because formula 1 is relevant with concrete network structure, therefore at above-mentioned three kinds of different network structures, can calculate numerical value F1, F2 and the F3 of three kinds of structures formula 1 separately respectively.The data of selected value maximum from these three data, for example F2 then is defined as the computational analysis network with the corresponding network structure of F2.When node data was more, training process was appreciated that with above-mentioned consistent the network structure that needs to select can corresponding increasing.
In step 303, according to the user preference network model of setting up and training, determine user's preference data correction threshold.For example, can be with the number of times substitution threshold calculations phase-split network of user's discussion with the relevant word of business tine preference, the preference of calculating the user is the probability P (preference i) of preference i, and conditional probability P (preference i threshold value | preference i).Conditional probability P (preference i threshold value | preference i) be illustrated under the situation of given preference i, preference i threshold value obtains the probability of certain numerical value.For example, when the preference probability of supposing the user satisfies under the situation of Di Li Cray probability density distribution, according to Di Li Cray probability density formula, the conditional probability P of each node i that can directly calculate (preference i threshold value | preference i).
According to resulting conditional probability value, calculate the scope of preference data correction threshold.
Wherein, τ
(preference i threshold value | preference i)Preference data correction threshold scope for preference i.
The preference data correction threshold scope of determining by said method has represented the zone of reasonableness of preference data correction threshold, and just the preference data correction threshold can be taken as any number in this scope.For example, can be according to requirement of client, required propelling movement precision is set at τ with the data correction threshold value of preference i
(preference i threshold value | preference i)The upper limit, lower limit, average etc.
In step 304, can compare by the number of times of the user being discussed the preference keyword that occurs in the record and the preference data correction threshold of determining, revise user's initial service content preference data.Thereby can will push to the user with the corresponding data service of revised business tine preference data.
In step 305, can be further active user's propelling movement result be fed back in the step 302, thereby further the user preference network model is trained, improve the accuracy of network model, thereby can accurately set the preference data correction threshold, further accurately to determine the data service of user's actual needs, improve the success ratio that pushes.
According to another aspect of the application, provide a kind of supplying system of data service.As shown in Figure 4, this supplying system 1 comprises determination module 10, processing module 20, and push module 30.
According to a kind of embodiment, processing module 20 can comprise discusses platform 21 and administration module 22.
Push module 30 and will push to the user with the corresponding data service of revised business tine preference data.
According to a kind of embodiment, the preference data correction threshold can be predefined.For example, threshold value setting module 2220 can preestablish the preference data correction threshold according to user's essential information and to pushing the requirement of success ratio.
According to another kind of embodiment, threshold value setting module 2220 can be set up the user preference network model, and determines user's preference data correction threshold according to the network model of setting up.As shown in Figure 5, threshold value setting module 2220 may further include model building module 2221, threshold determination module 2222.
According to a kind of embodiment, correcting module 2220 can further include feedback module 2223, active user's propelling movement result can be fed back to model building module 2221, thereby further the user preference network model be trained, improve the accuracy of network model.
Though the application utilizes foregoing description and embodiment to specify, the application is not so limited.The application's protection domain is limited by the claim in the appended claims, anyly is equal to replacement to what technical characterictic in the claim carried out, all should belong to the application institute restricted portion.
Claims (12)
1. the method for pushing of a data service comprises:
Determine user's initial service content preference data according to user basic information;
Initial service content preference data according to the user is set the discussion theme for the user, and invites the user to add discussion;
According to user's discussion record and the correlativity of initial service content preference data, user's initial service content preference data is revised; And
To push to the user with the corresponding data service of revising of business tine preference data.
2. the method for claim 1, wherein described according to user's discussion record and the correlativity of initial service content preference data, the step that user's initial service content preference data is revised further comprises:
Extract and add up user's preference keyword according to user's discussion record;
Set the preference data correction threshold;
A number of times of the preference keyword that extracts and add up and the preference data correction threshold that sets are compared, and revise professional preference data according to result relatively.
3. method as claimed in claim 2, wherein, the step of described setting preference data correction threshold comprises:
Determine the information of known users and push the result, and according to determined user's information and propelling movement situation, set up also training user preference network model; And
According to the user preference network model of setting up and training, determine user's preference data correction threshold.
4. method as claimed in claim 2, wherein, the step of described setting preference data correction threshold further comprises:
Feed back active user's propelling movement result, and further train the user preference network model according to the active user's who feeds back propelling movement result.
5. the method for claim 1, wherein described user basic information comprises user's Back ground Information, business conduct data and user's internet data.
6. the method for claim 1, wherein, described initial service content preference data according to the user is set the discussion theme for the user, and the step of inviting the user to add discussion comprises: according to the initial service content preference data user is carried out the preference classification, set the discussion theme according to the preference classification results for the user, and invite the user to add corresponding discussion theme.
7. the supplying system of a data service comprises:
Determination module is determined user's initial service content preference data according to user basic information;
Processing module, initial service content preference data according to the user is set the discussion theme for the user, invite the user to add discussion, according to user's discussion record and the correlativity of initial service content preference data, user's initial service content preference data is revised; And
Push module, will push to the user with the corresponding data service of revising of business tine preference data.
8. system as claimed in claim 7, wherein, described determination module comprises:
The information acquisition module obtains user base data, customer service behavioral data and user's internet data; And
The information analysis module is determined user's initial service content preference data by user base data, customer service behavioral data and the user's internet data that obtains.
9. system as claimed in claim 7, wherein, described processing module comprises:
Platform is discussed, according to the initial service content preference data user is carried out the preference classification, set the discussion theme according to the preference classification results for the user, and invite the user to add corresponding discussion theme; And
Administration module can be revised user's initial service content preference data according to user's discussion record and the correlativity between the initial service content preference data.
10. system as claimed in claim 9, wherein, described administration module comprises:
Statistical module is according to user's discussion record extraction and statistics user's preference keyword;
The threshold value setting module is set the preference data correction threshold;
Correcting module with the number of times of institute's preference keyword of extracting and add up and set the preference data correction threshold and compare, and is revised professional preference data according to the result of comparison.
11. system as claimed in claim 10, wherein, described threshold value setting module comprises:
Model building module is determined the information of known users and is pushed the result, and according to determined user's information and propelling movement situation, sets up also training user preference network model; And
Threshold determination module according to the user preference network model of setting up and training, is determined user's preference data correction threshold.
12. system as claimed in claim 11, wherein, described threshold value setting module further comprises:
Feedback module feeds back active user's propelling movement result, and further trains the user preference network model according to the active user's who feeds back propelling movement result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110452013.9A CN103186586B (en) | 2011-12-30 | The method for pushing of data service and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110452013.9A CN103186586B (en) | 2011-12-30 | The method for pushing of data service and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103186586A true CN103186586A (en) | 2013-07-03 |
| CN103186586B CN103186586B (en) | 2016-12-14 |
Family
ID=
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809584A (en) * | 2015-05-06 | 2015-07-29 | 中国南方电网有限责任公司电网技术研究中心 | Substation routing maintenance method and system |
| CN107170452A (en) * | 2017-04-27 | 2017-09-15 | 广东小天才科技有限公司 | The Adding Way and device of a kind of electronic meeting |
| CN107240019A (en) * | 2016-03-28 | 2017-10-10 | 阿里巴巴集团控股有限公司 | Assess customer service preference methods, customer investment risk partiality method and device |
| CN108959319A (en) * | 2017-05-25 | 2018-12-07 | 腾讯科技（深圳）有限公司 | Information-pushing method and device |
| CN112667887A (en) * | 2020-12-22 | 2021-04-16 | 北京达佳互联信息技术有限公司 | Content recommendation method and device, electronic equipment and server |
| CN113553105A (en) * | 2020-04-23 | 2021-10-26 | 百度在线网络技术（北京）有限公司 | Method and device for generating guide page |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101079824A (en) * | 2006-06-15 | 2007-11-28 | 腾讯科技（深圳）有限公司 | A generation system and method for user interest preference vector |
| US20110184977A1 (en) * | 2008-09-27 | 2011-07-28 | Jiachun Du | Recommendation method and system based on collaborative filtering |
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101079824A (en) * | 2006-06-15 | 2007-11-28 | 腾讯科技（深圳）有限公司 | A generation system and method for user interest preference vector |
| US20110184977A1 (en) * | 2008-09-27 | 2011-07-28 | Jiachun Du | Recommendation method and system based on collaborative filtering |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809584A (en) * | 2015-05-06 | 2015-07-29 | 中国南方电网有限责任公司电网技术研究中心 | Substation routing maintenance method and system |
| CN104809584B (en) * | 2015-05-06 | 2018-05-04 | 中国南方电网有限责任公司电网技术研究中心 | Substation patrols dimension method and system |
| CN107240019A (en) * | 2016-03-28 | 2017-10-10 | 阿里巴巴集团控股有限公司 | Assess customer service preference methods, customer investment risk partiality method and device |
| CN107170452A (en) * | 2017-04-27 | 2017-09-15 | 广东小天才科技有限公司 | The Adding Way and device of a kind of electronic meeting |
| CN108959319A (en) * | 2017-05-25 | 2018-12-07 | 腾讯科技（深圳）有限公司 | Information-pushing method and device |
| CN113553105A (en) * | 2020-04-23 | 2021-10-26 | 百度在线网络技术（北京）有限公司 | Method and device for generating guide page |
| CN112667887A (en) * | 2020-12-22 | 2021-04-16 | 北京达佳互联信息技术有限公司 | Content recommendation method and device, electronic equipment and server |
| CN112667887B (en) * | 2020-12-22 | 2024-03-12 | 北京达佳互联信息技术有限公司 | Content recommendation method and device, electronic equipment and server |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109428928B (en) | Method, device and equipment for selecting information push object | |
| CN107330020B (en) | User entity analysis method based on structure and attribute similarity | |
| CN103678613B (en) | Method and device for calculating influence data | |
| WO2015139559A1 (en) | Method and system for generating digital human | |
| CN105045916A (en) | Mobile game recommendation system and recommendation method thereof | |
| CN105447147B (en) | A kind of data processing method and device | |
| US20150081725A1 (en) | System and method for actively obtaining social data | |
| CN104657372A (en) | Page operation data processing method and device | |
| CN102695121A (en) | Method and system for pushing friend information for user in social network | |
| CN106529714A (en) | Method and system predicting user loss | |
| CN108763502A (en) | Information recommendation method and system | |
| CN106530015A (en) | Information releasing control method and apparatus | |
| CN105868267B (en) | A kind of modeling method of mobile social networking user interest | |
| CN105183909B (en) | social network user interest predicting method based on Gaussian mixture model | |
| CN102567534B (en) | Interactive product user generated content intercepting system and intercepting method for the same | |
| CN106127535A (en) | Advertisement push system based on mobile terminal and method | |
| CN105931079A (en) | Method and apparatus for diffusing seed users | |
| CN102811207A (en) | Network information pushing method and system | |
| CN108509545B (en) | Method and system for processing comments of article | |
| CN108090787A (en) | A kind of call bill data depth based on Apriori algorithm is excavated and the method for user's behavior prediction | |
| Ackland et al. | Political homophily on the web | |
| CN102739716B (en) | User information issuing method and server | |
| CN103186586A (en) | Pushing method and system for data service | |
| CN106570003B (en) | Data pushing method and device | |
| CN103186586B (en) | The method for pushing of data service and system |
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 |