CN103177129B - Internet real-time information recommendation prognoses system - Google Patents
Internet real-time information recommendation prognoses system Download PDFInfo
- Publication number
- CN103177129B CN103177129B CN201310139138.5A CN201310139138A CN103177129B CN 103177129 B CN103177129 B CN 103177129B CN 201310139138 A CN201310139138 A CN 201310139138A CN 103177129 B CN103177129 B CN 103177129B
- Authority
- CN
- China
- Prior art keywords
- information
- user
- unit
- internet
- real
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A kind of real-time information recommendation prognoses system of Internet technical field, comprise: data processing server and in real time application server, wherein: data processing server training is from each user and the data message browsing user, and real-time application server is browsed the internet information characteristic of user when displaying live view website or exported correlation analysis result to user; Real-time application server comprises: Internet user's recognition unit, real-time information release processing unit and Information issued prediction recommendation results output unit; Data processing server comprises: input block, storage unit, CPU (central processing unit) and output unit.The present invention can Realization analysis obtain other users and throw in the information in internet information, what utilize these information obtained can instruct user in real-time information issuing process takes corresponding strategy, these strategies include but not limited to instruct user to carry out guidance for user to carry out internet information issue, instruct user to provide various reports analysis etc.
Description
Technical field
What the present invention relates to is a kind of system of Internet technical field, specifically a kind of Internet real-time information recommendation prognoses system.
Background technology
Along with the high speed development of network, network, as maximum information carrier and intercommunion platform, has become current important channel of carrying out information propagating.Conventional information (as news, advertisement, commodity etc.) formulates Media Strategy, through position of media coverage goal audient, for realizing precisely, goes checking, adjustment more only after input with data, is difficult to realize considering that whether audient is interested and accept.And (as individual character news briefing, Instant Ads is bidded, individual character commodity displaying etc.) is pushed for Internet real-time information, before input, just search out suitable audient, really realize the accurate input of interconnected networking advertisement, individual character marketing.
In general, Information issued client can come by the attributive analysis intending releasing news to carry out contrasting with other Information issued client the Information issued space finally obtained from media.Such as, for real time bid ad system, party in request's platform, according to click-through rate, key word, the amount of money, advertisement and propaganda activity, carries out bidding with other users etc. index the advertising space finally obtained from media.
But Information issued client can obtain information in this Information issued process, can only by predicting from the historical information of throwing in internet information as user.Can not effectively profit from the information of other Information issued client, the blindness of information be caused to be issued.
The domestic and international research to internet information recommendation comprehensive analysis is also few at present, in existing information is recommended, existing utilizes internet information to carry out the technology recommended and predict, as Chinese patent literature CN101923545B, authorize day 2012-10-10, disclose a kind of method of recommendation of personalized information, suitable wireless Internet, behavior record module record on mobile phone, analyzes the content of the page of user's access, obtains the interested page set of user and user to the interest-degree of the page; The interest-degree of described interested page set and the page is combined, sets up user interest model, produce the customized information recommended.This technology can record analysis user behavior automatically, and digging user attribute and preference, achieve the coupling of user property and the information content, and initiatively recommend user.But the weak point of this technology compared with the present invention is: although this technology can be analyzed the behavior of user, but be only analyze for the behavior record of user self, other users that there is incidence relation with it can not be analyzed simultaneously, and and then oppositely recommendation and prediction are made to this user.
In existing information is recommended, also existing utilizes Internet user's relation to carry out the technology recommended and predict, as Chinese patent literature CN101540739B, authorizes day 2011.10.05, disclose a kind of user's recommend method, according to friend relation neural network forecast candidate good friend; Candidate good friend is returned user side.Above-mentioned user's recommend method is predicted according to the friend relation between user, does not rely on the personal information of user, and can expand communication face by friend relation.But the weak point of this technology compared with the present invention is: although this technology can utilize the behavior of friend relation to user to analyze, but only refer to already present friend relation network, a large amount of unknown subscriber's relation existed in interconnected Information issued process can not be utilized, and recommendation and prediction are made to this user.
In sum, need a kind of method and can consider internet browsing user behavior, user behavior realizes the recommendation prediction of comprehensive real-time Internet information thus tutorial message issue user carries out effective Information issued.
Summary of the invention
The present invention is directed to prior art above shortcomings, a kind of Internet real-time information recommendation prognoses system is proposed, can Realization analysis obtain other users and throw in the information in internet information, what utilize these information obtained can instruct user in real-time information issuing process takes corresponding strategy, these strategies include but not limited to instruct user to carry out guidance for user to carry out internet information issue, instruct user to provide various reports analysis etc.
The present invention is achieved by the following technical solutions, the present invention includes: data processing server and in real time application server, wherein: data processing server training is from each user and the data message browsing user, and real-time application server is browsed the internet information characteristic of user when displaying live view website or exported correlation analysis result to user.
Described real-time application server comprises: Internet user's recognition unit, real-time information release processing unit and Information issued prediction recommendation results output unit, wherein: the user that Internet user's recognition unit orientation logged in the past, real-time information release processing unit extracts corresponding prediction recommendation results according to the user that Internet user's recognition unit identifies and processes accordingly from the training result of data processing server.Final prediction recommendation results then exports by prediction recommendation results output unit.
Described data processing server comprises: input block, storage unit, CPU (central processing unit) and output unit, wherein: the output terminal of input block is connected with CPU (central processing unit) and transmits the networks congestion control information collected, CPU (central processing unit) generates decision tree and training set according to networks congestion control information, and is predicted the outcome by output unit with acoustooptic output user behavior analysis, preserve decision tree corresponding to user and training set by storage unit.
Described networks congestion control information comprises: the data that release news from the internet of website, internet information content data, internet information characteristic history data, user profile, browse user profile data.
Described input block is adopted and constantly reads network log by reptile, directly inputs from other databases, carries out text event detection from keyboard, carried out Speech input or carry out image input by scanner by microphone.
Described CPU (central processing unit) comprises: real-time signal analysis unit and forecast analysis unit, wherein: real-time signal analysis unit receive networks congestion control information from input block and obtain Information issued characteristic attribute by parsing, internet releases news, internet information content data, generate corresponding decision tree according to Information issued characteristic attribute wherein respectively and export storage unit and forecast analysis unit respectively to, carrying out user behavior analysis and generation forecast training set according to internet information content data wherein; Forecast analysis unit is according to decision tree and predict that training set is to carrying out decision tree training and user's behavior prediction.
Described Information issued characteristic attribute comprises: audient's classification, information category, message length, information mood, Information issued position, issue conclusion.
Described decision tree refers to: for each Information issued user, attribute for each Information issued position correspondence comprises: for Information issued user Si, there are audient's category attribute, information category attribute, message length attribute, information mood attribute, Information issued position attribution and issue conclusion attribute, and set up according to above-mentioned information and train the decision tree obtaining the unique correspondence of different user.
Described real-time signal analysis unit comprises: information detecting unit, real-time information characteristic analysis unit, internet release news analytic unit, the content analysis unit that releases news, customer analysis unit and data storage cell, wherein: information detecting unit detects and identifies and to release news for same internet, and different time carries out the result of successful information issue by different users; Real-time information characteristic analysis unit predicts that each user releases news at the network behavior of different time to same internet; The internet analytic unit that releases news obtains internet and to release news attribute information itself; The content information analysis unit that releases news obtain user throw in the attribute information of internet information; The data that data storage cell stores real-time information characteristic analysis unit, internet releases news analytic unit, the content analysis unit that releases news, customer analysis unit, customer analysis unit generate.
Described customer analysis unit comprises user's directed information analytic unit and user behavior analysis unit, wherein: user's directed information analytic unit determines the attribute information browsing or click the user released news; The behavior of the directed unique user of user behavior analysis unit, and according to the analytical behavior of unique user, obtain the analytical behavior of different classes of user.
Release news attribute information of described internet itself comprises: the size that internet releases news, size, position.
Described user throw in internet information attribute information comprise: the material throwing in internet information, internet information color, internet information classification.
Described user's directed information analytic unit obtains each Information issued and successfully throws in user profile, such as class of subscriber, and userbase, releases news the release time, and release news input location information.
Described forecast analysis unit comprises: Information issued user in predicting unit, behavior prediction unit and prediction rule storehouse, wherein:
The rule prediction prestored in the information that Information issued user in predicting unit is issued according to history and prediction rule storehouse obtains issuing the user of this information, and behavior prediction unit predicts the behavior that may release news according to the rule prestored in prediction rule storehouse.
Described real-time information release processing unit comprises: self-information issues predicting unit, other Information issued user in predicting unit, perpetual object database and Information issued result integrated unit, wherein: self-information is issued predicting unit and predicted according to the historical behavior of this Information issued user, generation forecast result; Other Information issued user in predicting unit are predicted other user behaviors according to this Information issued user, generating recommendations result; User sets up perpetual object database in advance according to the service logic of self, the user exported when other Information issued user in predicting unit is present in the database of perpetual object, then carry out Information issued forecast analysis to this user, prediction obtains the recommendation results of this user; Self-information is issued to predict the outcome to issue with other user profile to carry out merging and is exported final real-time information and issue result by Information issued result integrated unit.
Technique effect
Compared with prior art, the present invention can carry out automatic examination and analysb according to the internet information characteristic information of the different user of input; The size that can release news according to internet, size, position to release news attribute analysis to internet; Can to the input content that releases news, material, internet information color, internet information classification is analyzed; Can according to user transmit different information and carry out orientation to browsing the user released news, and the behavior of user to be analyzed, and can classify to user according to user behavior; Can predict possible Information issued according to the information of different user; Prediction can be issued according to carry out real-time information for different users and the potential different user of possibility.
Accompanying drawing explanation
Fig. 1 is the schematic diagram that real-time recommendation of the present invention predicts the device that releases news.
Fig. 2 is the schematic diagram of data processing server of the present invention.
Fig. 3 is the schematic diagram of real-time signal analysis unit of the present invention.
Fig. 4 is the schematic diagram of forecast analysis unit of the present invention.
Fig. 5 is the exemplary plot of Information issued forecast and decision of the present invention tree.
Fig. 6 is the schematic diagram of the real-time application server of the present invention.
Fig. 7 is the schematic diagram of real-time information release processing unit of the present invention.
Embodiment
Elaborate to embodiments of the invention below, the present embodiment is implemented under premised on technical solution of the present invention, give detailed embodiment and concrete operating process, but protection scope of the present invention is not limited to following embodiment.
Embodiment 1
As shown in Figure 1, the present embodiment device comprises: data processing server 101 and in real time application server 102, wherein: data processing server 101 is trained from each user and the data message browsing user, real-time application server 102 is browsed the internet information characteristic of user when displaying live view website or is exported correlation analysis result to user.
As shown in Figure 2, described data processing server 101 comprises: input block 201, storage unit 202, CPU (central processing unit) 203 and the output unit be made up of display unit 204 and voice unit (VU) 205, wherein: the output terminal of input block 201 is connected with CPU (central processing unit) 203 and transmits the networks congestion control information collected, CPU (central processing unit) 203 generates decision tree and training set according to networks congestion control information, and predicted the outcome with acoustooptic output user behavior analysis by display unit 204 and voice unit (VU) 205, decision tree corresponding to user and training set is preserved by storage unit 202.
Described networks congestion control information comprises: the data that release news from the internet of website, internet information content data, internet information characteristic history data, user profile, browse user profile data.
Described input block 101 can be various ways, constantly network log can be read by reptile, directly can input from other databases, text event detection can be carried out, also can carry out Speech input by microphone or carry out image input by scanner by keyboard.
As shown in Figure 3, described CPU (central processing unit) 203 comprises: real-time signal analysis unit 301 and forecast analysis unit 302, wherein: real-time signal analysis unit 301 receives networks congestion control information from input block 201 and obtains Information issued characteristic attribute by parsing, internet releases news, internet information content data, generate corresponding decision tree according to Information issued characteristic attribute wherein respectively and export storage unit 202 and forecast analysis unit 302 respectively to, user behavior analysis is carried out and generation forecast training set according to internet information content data wherein, forecast analysis unit 302 is according to decision tree and predict that training set is to carrying out decision tree training and user's behavior prediction.
Described Information issued characteristic attribute comprises: audient's classification, information category, message length, information mood, Information issued position, issue conclusion.
Described decision tree refers to: for each Information issued user, for each Information issued position, comprises a lot of attribute, such as: for Information issued user Si, there is audient's classification, information category, message length, information mood, Information issued position, issue the attributes such as conclusion:
Can set up according to above-mentioned information and train the decision tree for different user.
As shown in Figure 3, described real-time signal analysis unit comprises: information detecting unit 401, real-time information characteristic analysis unit 402, internet release news analytic unit 403, the content analysis unit that releases news 404, customer analysis unit 405 and data storage cell 406, wherein: information detecting unit 401 detects and identifies and to release news for same internet, and different time carries out the result of successful information issue by different users; Real-time information characteristic analysis unit 402 predicts that each user releases news at the network behavior of different time to same internet; The internet analytic unit 403 that releases news obtains internet and to release news attribute information itself; The content information analysis unit that releases news 404 obtain user throw in the attribute information of internet information; Customer analysis unit 405 comprises user's directed information analytic unit and user behavior analysis unit, wherein: user's directed information analytic unit determines the attribute information browsing or click the user released news, such as place, weather, temperature and the information such as user preferences, class of subscriber; User behavior analysis unit not only can the behavior of directed unique user, and according to the analytical behavior of unique user, can obtain the analytical behavior of different classes of user; The data that data storage cell 406 stores real-time information characteristic analysis unit, internet releases news analytic unit, the content analysis unit that releases news, customer analysis unit and customer analysis unit generate.
Release news attribute information of described internet itself comprises: the size that internet releases news, size, position etc.
Described user throw in internet information attribute information comprise: the material throwing in internet information, internet information color, internet information classification etc.
Described user's directed information analytic unit obtains each Information issued and successfully throws in user profile, such as class of subscriber, and userbase, releases news the release time, and release news information such as throwing in place.
Described user behavior analysis unit not only can the behavior of directed single user, and according to the analytical behavior of single user, can obtain the analytical behavior of different classes of user.
Illustrate, table 1 represents that certain Information issued business is its user A and releases news the infomation detection carried out for particular interconnect net, and within the time of T1 to Tn, user A and other different users release news for this internet and bid.This internet release news respectively take by different users (1 represent Information issued success, 0 represent Information issued failure).
Table 1
Described prediction refers to: the content that releases news (comprise and whether releasing news, and the relevant attribute released news is as information category, message length, information user, information mood etc.) predicted for user A, and concrete steps are as follows:
No user is detected to same Information issued position and issues situation, detected vector as follows:
…
Wherein: the sequence that each vector representation Information issued user released news in certain time, 1 expression releases news successfully, and 0 represents unsuccessful issue.The subscript of S represents Information issued position, and subscript represents different users.
represent Information issued user i Information issued position k, the situation that releases news of different time.With S
ithe immediate user of behavior is S
t:
S is found by similarity measurement
itop-N similar users
for with S
isimilar top-N user
with S
idecision tree be recorded to the storage unit of corresponding Si in the lump; By utilizing S
iwith analogous relationship user
decision tree obtain result for (c1, c2 ...., cn), wherein: ci be 1 or 0, i be arbitrary in 1 ~ n; The quantity that quantity when obtain 1 is greater than 0, then to S
irecommend the conclusion that impression information is such; Otherwise, then to S
irecommend the conclusion of not impression information.
Such as in table 2, user locations, sex, concern internet information classification, shopping hobby attribute information is shown, wherein: numeral is the numeral after each attribute discretization.Can obtain { user 1, and user 2 } through carrying out collaborative filtering classification to these certain customers, { user 3, and user 8 }, { user 4, and user 5, and user 7, and user 11, and user 12 }.Illustrate that { user 1, and user 2 } can be a classification, their behavior is comparatively close.{ user 3, and user 8 }, { user 4, and user 5, and user 7, and user 11, and user 12 } are respectively as two different classifications.Above-mentioned sorting technique is not limited to collaborative filtering, and the methods such as traditional cluster analysis can be suitable for completely.
Table 2
Described user's directed information analytic unit obtains each Information issued and successfully throws in user profile, such as class of subscriber, and userbase, releases news the release time, and release news information such as throwing in place.
Described user behavior analysis unit not only can the behavior of directed single user, and according to the analytical behavior of single user, can obtain the analytical behavior of different classes of user.
Such as in table 3, show class of subscriber, userbase, release news the release time, release news and throw in place, wherein: numeral is the numeral after each attribute discretization.Can obtain { user 1, and user 2 } through carrying out collaborative filtering classification to these certain customers, { user 3, and user 8 }, { user 4, and user 5, and user 7, and user 11, and user 12 }.Illustrate that { user 1, and user 2 } can be a classification, their behavior is comparatively close.{ user 3, and user 8 }, { user 4, and user 5, and user 7, and user 11, and user 12 } are respectively as two different classifications.Above-mentioned sorting technique is not limited to collaborative filtering, and the methods such as traditional cluster analysis can be suitable for completely.
Table 3
Described forecast analysis unit 302 comprises: user in predicting unit 501, behavior prediction unit 502 and prediction rule storehouse, wherein: data storage cell 406 stores the prediction training set that real-time signal analysis unit 301 generates.According to the user data that data storage cell 406 provides, carry out process by user in predicting unit 501 and obtain user in predicting rule, and be stored to prediction rule storehouse 503.According to the price data that data storage cell 406 provides, carry out process by behavior prediction unit 502 and obtain price expectation rule, and be stored to prediction rule storehouse 503.
As shown in Figure 4, described forecast analysis unit 302 comprises: Information issued user in predicting unit 501, behavior prediction unit 502 and prediction rule storehouse 503, wherein: the rule prediction prestored in the information that Information issued user in predicting unit 501 is issued according to history and prediction rule storehouse 503 obtains issuing the user of this information, and behavior prediction unit 502 predicts the behavior that may release news according to the rule prestored in prediction rule storehouse 503.
The historical data being input as Information issued of the present invention as shown in Table A 4.Here with the method for decision tree, the data analysis in his-and-hers watches A4.
Table A 4
Such as, for 4 conditional attribute indexs, category attribute is Information issued user, total n historical data:
(T
1,A
1,B
1,C
1,P
1)
(T
2,A
2,B
2,C
2,P
2)
…
(T
n,A
n,B
n,C
n,,P
n)
Usually due to the different in kind of each attribute, scale is different, for avoiding Sparse, often needs to carry out sliding-model control, simple carries out the new sequence of discretize to each attribute according to homogeneous space here,
(T
1′,A
1′,B
1′,C
1′,P
1)
(T
2′,A′
2,B′
2,C
2′,P
2)
…
(T
n′,A′
n,B′
n,C
n′,P
n)
To above-mentioned data construct decision tree, Fig. 5 shows the part generating decision tree.
When sending out present time T
1, clicking rate is A
2, internet releases news and is set to B
3, key word is C
9, first discretize is carried out to this attribute, obtains T
1', A '
2, B '
3, C
9'.By T
1', A '
2, B '
3, C
9' input in A5 decision tree, obtain
, therefore finally determine at time T
1, clicking rate is A
2, internet releases news and is set to B
3, key word is C
9user be P
5.
As shown in Figure 6, described real-time application server 102 comprises: Internet user's recognition unit 701, real-time information release processing unit 702 and Information issued prediction recommendation results output unit 703, wherein: the user that Internet user's recognition unit 701 orientation logged in the past, real-time information release processing unit 702 extracts corresponding prediction recommendation results according to the user that Internet user's recognition unit 701 identifies and processes accordingly from the training result the storage unit 202 of data processing server 101, final prediction recommendation results exports by prediction recommendation results output unit 703.
As shown in Figure 7, described real-time information release processing unit 702 comprises: self-information issues predicting unit 801, other Information issued user in predicting unit 802, perpetual object database 803 and Information issued result integrated unit 804, wherein: self-information is issued predicting unit 801 and predicted according to the historical behavior of this Information issued user, generation forecast result; Other Information issued user in predicting unit 802 are predicted other user behaviors according to this Information issued user, generating recommendations result; User sets up perpetual object database 803 in advance according to the service logic of self, the user exported when other Information issued user in predicting unit 802 is present in the database 803 of perpetual object, then carry out Information issued forecast analysis to this user, prediction obtains the recommendation results of this user; Self-information is issued to predict the outcome to issue with other user profile to carry out merging and is exported final real-time information and issue result 804 by Information issued result integrated unit 804.
Such as: user A carries out self-information according to the input attributes that releases news and issues and predict the outcome as issuing successfully, predict that user B will be at war with Information issued by other Information issued user in predicting unit.The database of the rival of user A is arranged in due to user B.Other user profile is issued predicting unit and is exported the B also possibility information briefing success that releases news, for avoiding releasing news in this internet the chance losing and release news, user A will release news (as information category to original plan, message length, information price etc.) modify and again predict until prediction user A issues successfully, and user B issues unsuccessful.
Claims (6)
1. a real-time information recommendation prognoses system, it is characterized in that, comprise: data processing server and in real time application server, wherein: data processing server training is from each user and the data message browsing user, and real-time application server is browsed the internet information characteristic of user when displaying live view website or exported correlation analysis result to user;
Described real-time application server comprises: Internet user's recognition unit, real-time information release processing unit and Information issued prediction recommendation results output unit, wherein: the user that Internet user's recognition unit orientation logged in the past, real-time information release processing unit extracts corresponding prediction recommendation results according to the user that Internet user's recognition unit identifies and processes accordingly from the training result of data processing server
Final prediction recommendation results then exports by prediction recommendation results output unit;
Described data processing server comprises: input block, storage unit, CPU (central processing unit) and output unit, wherein: the output terminal of input block is connected with CPU (central processing unit) and transmits the networks congestion control information collected, CPU (central processing unit) generates decision tree and training set according to networks congestion control information, and is predicted the outcome by output unit with acoustooptic output user behavior analysis, preserve decision tree corresponding to user and training set by storage unit;
Described networks congestion control information comprises: the data that release news from the internet of website, internet information content data, internet information characteristic history data, user profile, browse user profile data;
Described CPU (central processing unit) comprises: real-time signal analysis unit and forecast analysis unit, wherein: real-time signal analysis unit receive networks congestion control information from input block and obtain Information issued characteristic attribute by parsing, internet releases news, internet information content data, generate corresponding decision tree according to Information issued characteristic attribute wherein respectively and export storage unit and forecast analysis unit respectively to, carrying out user behavior analysis and generation forecast training set according to internet information content data wherein; Forecast analysis unit is according to decision tree and predict that training set is to carrying out decision tree training and user's behavior prediction;
Described Information issued characteristic attribute comprises: audient's classification, information category, message length, information mood, Information issued position, issue conclusion.
2. system according to claim 1, it is characterized in that, described input block is adopted and constantly reads network log by reptile, directly inputs from other databases, carries out text event detection from keyboard, carried out Speech input or carry out image input by scanner by microphone.
3. system according to claim 1, it is characterized in that, described decision tree refers to: for each Information issued user, attribute for each Information issued position correspondence comprises: for Information issued user Si, there are audient's category attribute, information category attribute, message length attribute, information mood attribute, Information issued position attribution and issue conclusion attribute, and set up according to above-mentioned information and train the decision tree obtaining the unique correspondence of different user.
4. system according to claim 1, it is characterized in that, described real-time signal analysis unit comprises: information detecting unit, real-time information characteristic analysis unit, internet release news analytic unit, the content analysis unit that releases news, customer analysis unit and data storage cell, wherein: information detecting unit detects and identifies and to release news for same internet, and different time carries out the result of successful information issue by different users; Real-time information characteristic analysis unit predicts that each user releases news at the network behavior of different time to same internet; The internet analytic unit that releases news obtains internet and to release news attribute information itself; The content information analysis unit that releases news obtain user throw in the attribute information of internet information; The data that data storage cell stores real-time information characteristic analysis unit, internet releases news analytic unit, the content analysis unit that releases news, customer analysis unit, customer analysis unit generate;
Described customer analysis unit comprises user's directed information analytic unit and user behavior analysis unit, wherein: user's directed information analytic unit determines the attribute information browsing or click the user released news; The behavior of the directed unique user of user behavior analysis unit, and according to the analytical behavior of unique user, obtain the analytical behavior of different classes of user;
Release news attribute information of described internet itself comprises: the size that internet releases news, size, position;
Described user throw in internet information attribute information comprise: the material throwing in internet information, internet information color, internet information classification;
Described user's directed information analytic unit obtains each Information issued and successfully throws in user profile, such as class of subscriber, and userbase, releases news the release time, and release news input location information.
5. system according to claim 1, it is characterized in that, described forecast analysis unit comprises: Information issued user in predicting unit, behavior prediction unit and prediction rule storehouse, wherein: the rule prediction prestored in the information that Information issued user in predicting unit is issued according to history and prediction rule storehouse obtains issuing the user of this information, and behavior prediction unit predicts the behavior that may release news according to the rule prestored in prediction rule storehouse.
6. system according to claim 1, it is characterized in that, described real-time information release processing unit comprises: self-information issues predicting unit, other Information issued user in predicting unit, perpetual object database and Information issued result integrated unit, wherein: self-information is issued predicting unit and predicted according to the historical behavior of this Information issued user, generation forecast result; Other Information issued user in predicting unit are predicted other user behaviors according to this Information issued user, generating recommendations result; User sets up perpetual object database in advance according to the service logic of self, the user exported when other Information issued user in predicting unit is present in the database of perpetual object, then carry out Information issued forecast analysis to this user, prediction obtains the recommendation results of this user; Self-information is issued to predict the outcome to issue with other user profile to carry out merging and is exported final real-time information and issue result by Information issued result integrated unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310139138.5A CN103177129B (en) | 2013-04-19 | 2013-04-19 | Internet real-time information recommendation prognoses system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310139138.5A CN103177129B (en) | 2013-04-19 | 2013-04-19 | Internet real-time information recommendation prognoses system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103177129A CN103177129A (en) | 2013-06-26 |
| CN103177129B true CN103177129B (en) | 2016-03-16 |
Family
ID=48636990
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310139138.5A Active CN103177129B (en) | 2013-04-19 | 2013-04-19 | Internet real-time information recommendation prognoses system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103177129B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107688956A (en) * | 2016-08-05 | 2018-02-13 | 腾讯科技（深圳）有限公司 | A kind of information processing method and server |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104376027B (en) * | 2013-08-16 | 2018-07-10 | 上海未来宽带技术股份有限公司 | Information acquisition and analysis system and its method based on social media |
| CN103399965A (en) * | 2013-08-26 | 2013-11-20 | 百度在线网络技术（北京）有限公司 | Reading content recommending method, reading content recommending system and server |
| CN103473352A (en) * | 2013-09-24 | 2013-12-25 | 上海新数网络科技有限公司 | Integrated individualized information pushing system |
| CN104717124B (en) * | 2013-12-13 | 2019-07-16 | 腾讯科技（深圳）有限公司 | A kind of friend recommendation method, apparatus and server |
| CN103942707B8 (en) * | 2014-04-08 | 2018-06-29 | 璧合科技股份有限公司 | Advertising results optimization system based on real time bid |
| US11228653B2 (en) | 2014-05-15 | 2022-01-18 | Samsung Electronics Co., Ltd. | Terminal, cloud apparatus, driving method of terminal, method for processing cooperative data, computer readable recording medium |
| KR20150132800A (en) * | 2014-05-15 | 2015-11-26 | 삼성전자주식회사 | Terminal, Cloud Apparatus, Driving Method of Terminal, Method for Providing Cloud Service, Computer Readable Recording Medium |
| CN104618438A (en) * | 2014-12-29 | 2015-05-13 | 厦门欣旅通科技有限公司 | Content distribution method and device |
| CN108304164B (en) * | 2017-09-12 | 2021-12-03 | 马上消费金融股份有限公司 | Business logic development method and development system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101110699A (en) * | 2007-08-07 | 2008-01-23 | 广州诚予国际市场信息研究有限公司 | System with network satisfaction degree estimation and early warning function and implementing method thereof |
| CN101923545A (en) * | 2009-06-15 | 2010-12-22 | 北京百分通联传媒技术有限公司 | Method for recommending personalized information |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE60202021T2 (en) * | 2001-07-03 | 2005-11-10 | Ishida Europe Ltd. | distribution device |
-
2013
- 2013-04-19 CN CN201310139138.5A patent/CN103177129B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101110699A (en) * | 2007-08-07 | 2008-01-23 | 广州诚予国际市场信息研究有限公司 | System with network satisfaction degree estimation and early warning function and implementing method thereof |
| CN101923545A (en) * | 2009-06-15 | 2010-12-22 | 北京百分通联传媒技术有限公司 | Method for recommending personalized information |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107688956A (en) * | 2016-08-05 | 2018-02-13 | 腾讯科技（深圳）有限公司 | A kind of information processing method and server |
| CN107688956B (en) * | 2016-08-05 | 2021-09-07 | 腾讯科技（深圳）有限公司 | Information processing method and server |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103177129A (en) | 2013-06-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103177129B (en) | Internet real-time information recommendation prognoses system | |
| Fan et al. | Demystifying big data analytics for business intelligence through the lens of marketing mix | |
| EP3407285B1 (en) | Target user orientation method and device, and computer storage medium | |
| CN102708131B (en) | By consumer's automatic classification in fine point | |
| US20150006286A1 (en) | Targeting users based on categorical content interactions | |
| US20150006294A1 (en) | Targeting rules based on previous recommendations | |
| US20160140627A1 (en) | Generating high quality leads for marketing campaigns | |
| CN106997549A (en) | The method for pushing and system of a kind of advertising message | |
| KR102297669B1 (en) | System for providing matching service for connecting between manufacturer and distributor | |
| US20140006153A1 (en) | System for making personalized offers for business facilitation of an entity and methods thereof | |
| CN107636715A (en) | System and method for providing content | |
| CN104254851A (en) | Method and system for recommending content to a user | |
| US20190385181A1 (en) | Using media information for improving direct marketing response rate | |
| JP2016062411A (en) | Information analysis device, information analysis method, and information analysis program | |
| US20180285748A1 (en) | Performance metric prediction for delivery of electronic media content items | |
| Wang et al. | Online ad effectiveness evaluation with a two-stage method using a Gaussian filter and decision tree approach | |
| Frey et al. | Mobile recommendations based on interest prediction from consumer's installed apps–insights from a large-scale field study | |
| Guo et al. | Harnessing the power of the general public for crowdsourced business intelligence: a survey | |
| Zhang et al. | Event organization 101: Understanding latent factors of event popularity | |
| CN113052653A (en) | Financial product content recommendation method and system and computer readable storage medium | |
| Chen | Comparing content marketing strategies of digital brands using machine learning | |
| US10331713B1 (en) | User activity analysis using word clouds | |
| US20150106158A1 (en) | Method and apparatus for providing folksonomic object scoring | |
| KR20160130369A (en) | In online web text based commercial analysis service system and method thereof | |
| KR102238438B1 (en) | System for providing commercial product transaction service using price standardization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 201800 Shanghai City, Jiading District Nanxiang Town Jia good road No. 1690 Building 2 room 423 Applicant after: Shanghai WiseMedia Technology Co., Ltd. Address before: 201800 Shanghai City, Jiading District Nanxiang Town Jia good road No. 1690 Building 2 room 423 Applicant before: Shanghai WiseMedia Technology Co., Ltd. |
|
| COR | Change of bibliographic data | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |