CN101079063A - Method, system and apparatus for transmitting advertisement based on scene information - Google Patents
Method, system and apparatus for transmitting advertisement based on scene information Download PDFInfo
- Publication number
- CN101079063A CN101079063A CNA2007100761267A CN200710076126A CN101079063A CN 101079063 A CN101079063 A CN 101079063A CN A2007100761267 A CNA2007100761267 A CN A2007100761267A CN 200710076126 A CN200710076126 A CN 200710076126A CN 101079063 A CN101079063 A CN 101079063A
- Authority
- CN
- China
- Prior art keywords
- user
- scene information
- interest
- advertisement
- feature speech
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The invention provides a method, a system and an equipment of pushing the advertisement based on the spot information in the communication field, which comprises the following steps: A, proceeding statistic of user accessing web page and generating the interest vector according to the extracting characteristic word from the web page; B, forecasting user's interest according to the interest vector and generating the spot information with the interest strike; C, selecting the advertisement according to the spot information and pushing the advertisement to the client. The invention improves the placement of pushing the advertisement, which completes the process of pushing the advertisement.
Description
Technical field
The present invention relates to the communications field, more particularly, relate to a kind of method, system and equipment that pushes advertisement based on scene information.
Background technology
At this economy era of taking as the leading factor with information communication, perfect along with Internet technology, the network intelligence advertisement is also in fast development.
The core technology of network intelligence advertisement is to carry out audient's analysis.Also promptly, draw user's characteristic information by network behavior analysis to the Internet user, such as age of this user, sex, geographic position, income situation with and interested field etc.
One of them aspect that the audient analyzes is to carry out scene analysis, and scene information can be used as the foundation that the user of those impulsive style shopping is pushed advertisement.So-called scene analysis, it is exactly web page browsing behavior by the statistics user, as browsing histories, browse activity etc., thereby analyze and learn which field is user's interest (for example concentrate on, automobile, house property, tourism, number, music, animation, recreation, physical culture, friend-making, reading, military affairs, finance and economics, literature, cuisines etc.), then can throw in the user's interest personalized advertisement pointedly then.
The detailed process that prior art is carried out scene analysis comprises: (1) utilizes participle technique to extract the web page characteristics speech, and webpage is classified; (2) quantity of the webpage crossed of statistic of user accessing extracts the feature speech of all webpages, and with its union as the candidate feature speech; (3) calculate each feature speech and appear at weights in each page, thereby obtain interest vector; (4) generate scene information according to interest vector, and after to its coding, write among the Cookie.Its detailed process that generates scene information comprises: interest vector averaged, and regularization again (making the weight sum is 1), gained is the value of scene information.
As from the foregoing, the resulting scene information of prior art only be to one of user interest comprehensive, can't embody the user interest trend, therefore the advertisement that pushes can not very accurately coincide user's reality interest and need.So-called user interest trend is meant that the user's interest field may be at any time all in transfer, and its transfer tendency then is called as the user interest trend.The scene information of prior art is not considered this user interest trend, therefore still remain aspect the specific aim that pushes advertisement further perfect.
Summary of the invention
The object of the present invention is to provide a kind of system, be intended to solve the problem that prior art pushes the specific aim deficiency of advertisement based on scene information propelling movement advertisement.
The present invention also aims to provide a kind of equipment, to solve the above-mentioned problems in the prior art better based on scene information propelling movement advertisement.
The present invention also aims to provide a kind of method, to solve the above-mentioned problems in the prior art better based on scene information propelling movement advertisement.
In order to realize goal of the invention, described system based on scene information propelling movement advertisement comprises the server and client side, described server comprises a database that is used to store the webpage that the user visits, a webpage that is used for visiting according to the user obtains the scene analysis unit of scene information, with an advertisement pushing unit according to the scene information advertisement delivery, comprise in the described scene analysis unit that an interest moves towards prediction module, the webpage predictive user interest that is used for visiting according to the user is moved towards, and generates a scene information that comprises described user interest trend.
Preferably, described scene analysis unit also comprises feature speech extraction module, interest vector generation module;
The webpage that described feature speech extraction module is used for browsing from the user extracts the feature speech;
Described interest vector generation module links to each other with feature speech extraction module, is used for calculating the weights of described feature speech at each webpage that the user browsed, and interest vector is formed in described feature speech and weights thereof, and send into interest and move towards prediction module.
Preferably, described server further comprises an effect analysis unit, links to each other with described advertisement pushing unit, is used for the user click data according to advertisement pushing data and client feedback, calculates the exposure rate and the clicking rate of advertisement.
In order to realize goal of the invention better, the described equipment that pushes advertisement based on scene information, the i.e. server that links to each other with client, described server comprises that is used to collect the also database of storaging user data, a scene analysis unit that is used for obtaining scene information according to user data, with an advertisement pushing unit according to the scene information advertisement delivery, comprise in the described scene analysis unit that an interest moves towards prediction module, the webpage predictive user interest that is used for visiting according to the user is moved towards, and generates a scene information that comprises described user interest trend.
Preferably, described scene analysis unit also comprises feature speech extraction module, interest vector generation module;
The webpage that described feature speech extraction module is used for browsing from the user extracts the feature speech;
Described interest vector generation module links to each other with feature speech extraction module, is used for calculating the weights of described feature speech at each webpage that the user browsed, and interest vector is formed in described feature speech and weights thereof, and send into interest and move towards prediction module.
In order to realize goal of the invention better, described method based on scene information propelling movement advertisement may further comprise the steps:
A. statistic of user accessing web page, and generate interest vector according to the feature speech that from described webpage, extracts;
B. move towards according to described interest vector predictive user interest, and generate a scene information that comprises the user interest trend;
C. select advertisement according to scene information, and with described advertisement pushing to the client at user place.
Preferably, also comprise before the described steps A: utilize participle technique that webpage is classified.
Preferably, described steps A further comprises:
A1. from user accessing web page, extract the feature speech, and with the union of described feature speech as the candidate feature speech;
A2. weights of each in the calculated candidate feature speech, and feature speech and weights thereof be organized into an interest vector.
Preferably, described step B comprises:
B11. each component in the described interest vector is weighted on average, wherein the weight value of each component and its time are inversely proportional in proper order;
B12. the weighted mean value to step B11 output carries out the regularization processing, and operation result is encoded as scene information.
Preferably, described step B comprises:
B21. with described interest vector as the value of multidimensional time series on corresponding time point;
B22. predict the next value of described multidimensional seasonal effect in time series, and described next value is encoded as scene information.
The present invention obtains embodying the scene information of user interest trend by the nearest webpage of browsing of user is added up, and has effectively improved the specific aim of advertisement pushing; In addition, also field feedback is carried out effect analysis, thus further perfect advertisement pushing process.
Description of drawings
Fig. 1 is the system construction drawing that pushes advertisement among the present invention based on scene information;
Fig. 2 is the cut-away view of scene analytic unit in the server of the present invention;
Fig. 3 is the system construction drawing that pushes advertisement in one embodiment of the present of invention based on scene information;
Fig. 4 is the method flow diagram that pushes advertisement among the present invention based on scene information;
Fig. 5 is the method flow diagram that pushes advertisement in one embodiment of the present of invention based on scene information.
Embodiment
In order to make purpose of the present invention, technical scheme and advantage clearer,, the present invention is further elaborated below in conjunction with drawings and Examples.Should be appreciated that specific embodiment described herein only in order to explanation the present invention, and be not used in qualification the present invention.
Among the present invention, at first utilize participle technique to extract feature speech in all webpages, thereby webpage is classified; Regularly add up the webpage that the user visited recently then, the union of the feature speech in the webpage as the candidate feature speech, and is calculated each feature speech and appeared at weights in each webpage, thereby obtain interest vector; Further obtain a scene information that can embody the user interest trend according to interest vector again, thereby can push the client of advertisement pointedly according to this scene information to the user place.
Fig. 1 shows the system architecture that pushes advertisement among the present invention based on scene information, and this system comprises server 100, and coupled a plurality of clients (client 200, client 300 ... client N).Should be noted that the annexation between each equipment is the needs of explaining its information interaction and control procedure for clear in all diagrams of the present invention, therefore should be considered as annexation in logic, and should not only limit to physical connection.
Each client (client 200, client 300 ... client N) can be the various terminal devices that to login the Internet and carry out network activity; personal computer (Personal Computer for example; PC), personal digital assistant (Personal Digital Assistant; PDA), mobile phone (Mobile Phone; MP) etc., thus protection scope of the present invention should not be defined as the client of certain particular type.
(1) database 101 is used to store collected user data, and these user data mainly comprise browsing histories (being the webpage that the user browses) of user, user's the network behavior clicking operation of its interested link (for example to) etc.
(2) scene analysis unit 102 links to each other with database 101, be used for carrying out scene analysis according to database 101 storage user data, mainly be that the webpage that the user is browsed is added up, and calculating interest vector according to the feature speech that from webpage, extracts, this interest vector has reflected the user's interest field from numerical value.After generating this interest vector, again it is encoded, and write among the Cookie of webpage.Realize about the inner structure of this scene analysis unit 102 and concrete function, in Fig. 2, will be described in detail.
(3) advertisement pushing unit 103 links to each other with scene analysis unit 102, is used for according to the interest vector coding that writes webpage Cookie each client (client 200, client 300 being selected and pushed in advertisement ... client N).
Fig. 2 shows the inner structure of scene analytic unit 102 in the server 100 of Fig. 1, comprises that feature speech extraction module 1021, interest vector generation module 1022, interest moves towards prediction module 1023, wherein:
(1) feature speech extraction module 1021 is used for adding up the webpage that users visited recently from 101 storage user data of database, and therefrom extracts the feature speech.Among the present invention, feature speech extraction module 1021 can start statistical operation based on multiple condition, and its mode of extracting the feature speech also has multiple.
In an exemplary scenario that starts statistical operation, this feature speech extraction module 1021 is modes of taking timing extraction.Specific implementation is: a time threshold T is set in feature speech extraction module 1021, and then every the time T webpage that then the statistics user visited recently from 101 storage user data of database, webpage quantity to be added up also can be provided with flexibly.
In an exemplary scenario extracting the feature speech, feature speech extraction module 1021 is before extracting the feature speech, need earlier all webpages to be classified, specifically comprise: crucial dictionary of predefine, entry source wherein can be with reference to the searching key word of widespread use in network, and define the interest pattern under each entry, for example automobile, house property, tourism, number, music, animation, recreation, physical culture, friend-making, reading, military affairs, finance and economics, literature, cuisines or the like in this key dictionary; Take participle technique then, extract the feature term vector of webpage, calculate itself and the similarity of the corresponding keyword of each interest pattern, and webpage is referred to the interest class of similarity value maximum.Feature speech extraction module 1021 is based on this classification then, add up the quantity of the webpage that the user visited recently in each interest pattern, and adopting participle technique from the webpage of being visited, to extract the feature speech, the union of feature speech of obtaining all webpages again is as the candidate feature speech of user interest vector.In one embodiment, feature speech note is made K
i, the complete or collected works K={K of user interest category feature speech so
1, K
2..., K
m, wherein m represents the quantity of feature speech.
(2) interest vector generation module 1022 links to each other with feature speech extraction module 1021, is used for generating corresponding interest vector according to the feature speech that extracts.Specifically comprise: at first calculate the weights of all feature speech, then with the combination of all feature speech and its weights as interest vector.Wherein, the weight calculation method of feature speech comprises multiple.
In one embodiment, utilize the weights of following computing formula calculated characteristics speech:
Wherein, m is the quantity of feature speech among the aforementioned feature speech complete or collected works, and n is the webpage quantity of being added up, i=1, and 2 ..., m; J=1,2 ..., n.Tf
IjRepresentation feature speech K
iAt page P
jIn weights,
It is the weight of being got when calculating the weights of i feature speech in n the page.
After having calculated the weights of each feature speech successively, again with the combination of feature speech and its weights as interest vector, be designated as Uc, be expressed as: Uc={ (K1, W1) ..., (Ki, Wi) ..., (Km, Wm) }.
(3) interest is moved towards prediction module 1023 and is linked to each other with interest vector generation module 1022, be used for the trend of user interest being predicted according to interest vector, generate a scene information that comprises the user interest trend, and will write among the Cookie of webpage behind this scene information coding.Among the present invention, the scene information that comprises the user interest trend can have various ways, and its generating mode also has multiple.
In generating an exemplary scenario of scene information, interest vector is weighted average computing, the weight the closer to nearest interest vector is obtained big more, also, weight value and its time are inversely proportional in proper order; Then the result after the weighted mean is carried out regularization and handle (making the weight sum is 1), income value is scene information of the present invention.Because in this generation method, it is simple average that interest vector is had more than, but is weighted on average, the value of weight has been considered the drift of scene information, therefore can embody the user interest trend.The scene information that but utilizes this mode to generate remains an integrated value, is partial to remarkable inadequately.
In another exemplary scenario that generates scene information, then utilize the multidimensional time series to carry out computing.Specifically comprise: with described interest vector as the value of multidimensional time series on corresponding time point; Adopt data digging method to predict the next value of this multidimensional seasonal effect in time series then, and should next one value as scene information.In this exemplary scenario, be used to predict that the data digging method of the next value of multidimensional seasonal effect in time series can be multiple, (the Autoregressive IntegratedMoving Average of the autoregression moving average in the statistics for example, ARIMA), and any Forecasting Methodology in the machine learning (for example, neural network, decision tree, support vector machine etc.).
Fig. 3 shows the system architecture that pushes advertisement in one embodiment of the present of invention based on scene information, comprises server 100, and coupled a plurality of clients (client 200, client 300 ... client N).Compare with system shown in Figure 1, server 100 also comprises an effect analysis unit 104 except that comprising database 101, scene analysis unit 102 and advertisement pushing unit 103.
This effect analysis unit 104 links to each other with advertisement pushing unit 103, be used for advertisement pushing data and each client (client 200, client 300 according to advertisement pushing unit 103 records ... client N) Fan Kui user click data calculates the exposure rate and the clicking rate of advertisement.
Among the present invention, the computing method of exposure rate and hit rate can have multiple.In an exemplary scenario, the formula of effect analysis unit 104 calculation exposure rates is as follows: exposure rate=covering number of users/total number of users; Its formula that calculates hit rate is as follows: hit rate=clicks/impression.
In an embodiment of above-mentioned exemplary scenario, gained exposure rate and hit rate such as following table:
| The attribute classification | Number of users | Advertised name | Exposure frequency | Exposure rate | Number of clicks | Hit rate |
| The automobile group | 1,000,000 | BMW S series | 900,000 | 90% 300,000 | 33.3％ | |
| Women user | 5,000,000 | LUX beauty soap | 3,000,000 | 60% 2,400,000 | 80％ | |
In another embodiment of above-mentioned exemplary scenario, gained exposure rate and hit rate such as following table:
| The attribute classification | Number of users | Advertised name | Exposure frequency | Exposure rate | Number of clicks | Hit rate |
| 25～30 years old the male sex's white collar in Shenzhen | 1,000,000 | The sale advertising of sub-district, South Mountain | 400,000 | 40% 100,000 | 25％ | |
| Family income surpasses 500,000 the women more than 30 years old | 500,000 | The new product clothes | 400,000 | 80% 300,000 | 75％ | |
Certainly, effect analysis unit 104 also can calculate the exposure rate and the hit rate of advertisement by other modes among the present invention, is not limited to above-described method.
Fig. 4 shows and the present invention is based on the method flow that scene information pushes advertisement, and this method flow is based on Fig. 1, system architecture shown in Figure 2, and detailed process is as follows:
Carry out institute of the present invention in steps before, need earlier all webpages to be classified, specifically comprise: crucial dictionary of predefine, entry source wherein can be with reference to the searching key word of widespread use in network, and define the interest pattern under each entry, for example automobile, house property, tourism, number, music, animation, recreation, physical culture, friend-making, reading, military affairs, finance and economics, literature, cuisines or the like in this key dictionary; Take participle technique then, extract the feature term vector of webpage, calculate itself and the similarity of the corresponding keyword of each interest pattern, and webpage is referred to the interest class of similarity value maximum.
In step S401, server 100 statistic of user accessing web page are extracted the feature speech from webpage, and generate the user interest vector according to the feature speech.Specifically comprise: the quantity that (1) server 100 utilizes the feature speech extraction module 1021 in the scene analysis unit 102 to add up the webpage that the user visited recently in each interest pattern, and adopting participle technique from the webpage of being visited, to extract the feature speech, the union of feature speech of obtaining all webpages again is as the candidate feature speech of user interest vector; (2) interest vector generation module 1022 calculates the weights of all feature speech again, then with the combination of all feature speech and its weights as interest vector.
In one embodiment, feature speech note is made K
i, the complete or collected works K={K of user interest category feature speech so
1, K
2..., K
m, wherein m represents the quantity of feature speech.
And in this embodiment, can utilize the weights of following computing formula calculated characteristics speech:
Wherein, n is the webpage quantity of being added up, i=1, and 2 ..., m; J=1,2 ..., n.Tf
IjRepresentation feature speech K
iAt page P
jIn weights,
It is the weight of being got when calculating the weights of i feature speech in n the page.
After having calculated the weights of each feature speech successively, again with the combination of feature speech and its weights as interest vector, be designated as Uc, be expressed as: Uc={ (K1, W1) ..., (Ki, Wi) ..., (Km, Wm) }.
In step S402, server 100 utilizes the interest in its scene analysis unit 102 to move towards prediction module 1023, according to interest vector the trend of user interest is predicted, generate a scene information that comprises the user interest trend, and will write among the Cookie of webpage behind this scene information coding.Among the present invention, the scene information that comprises the user interest trend can have various ways, and its generating mode also has multiple.
In generating an exemplary scenario of scene information, interest vector is weighted average computing, the weight the closer to nearest interest vector is obtained big more, also, weight value and its time are inversely proportional in proper order; Then the result after the weighted mean is carried out regularization and handle (making the weight sum is 1), income value is scene information of the present invention.Because in this generation method, it is simple average that interest vector is had more than, but is weighted on average, the value of weight has been considered the drift of scene information, therefore can embody the user interest trend.The scene information that but utilizes this mode to generate remains an integrated value, is partial to remarkable inadequately.
In another exemplary scenario that generates scene information, then utilize the multidimensional time series to carry out computing.Specifically comprise: with described interest vector as the value of multidimensional time series on corresponding time point; Adopt data digging method to predict the next value of this multidimensional seasonal effect in time series then, and should next one value as scene information.In this exemplary scenario, be used to predict that the data digging method of the next value of multidimensional seasonal effect in time series can be multiple, (the Autoregressive IntegratedMoving Average of the autoregression moving average in the statistics for example, ARIMA), and any Forecasting Methodology in the machine learning (for example, neural network, decision tree, support vector machine etc.).
In step S403, select advertisement pointedly according to the scene information behind the coding, and selected advertisement pushing is given the client at user place.Owing to include the factor of user interest trend in this scene information, therefore the selected advertisement nearest demand of can being close to the users is carried out then and is thrown in operation.
Fig. 5 shows the method flow that pushes advertisement in one embodiment of the present of invention based on scene information, and this method flow is based on system architecture shown in Figure 3, and detailed process is as follows:
Carry out institute of the present invention in steps before, need earlier all webpages to be classified, specifically comprise: crucial dictionary of predefine, entry source wherein can be with reference to the searching key word of widespread use in network, and define the interest pattern under each entry, for example automobile, house property, tourism, number, music, animation, recreation, physical culture, friend-making, reading, military affairs, finance and economics, literature, cuisines or the like in this key dictionary; Take participle technique then, extract the feature term vector of webpage, calculate itself and the similarity of the corresponding keyword of each interest pattern, and webpage is referred to the interest class of similarity value maximum.
In step S501, server 100 statistic of user accessing web page are extracted the feature speech from webpage, and generate the user interest vector according to the feature speech.Specifically comprise: the quantity that (1) server 100 utilizes the feature speech extraction module 1021 in the scene analysis unit 102 to add up the webpage that the user visited recently in each interest pattern, and adopting participle technique from the webpage of being visited, to extract the feature speech, the union of feature speech of obtaining all webpages again is as the candidate feature speech of user interest vector; (2) interest vector generation module 1022 calculates the weights of all feature speech again, then with the combination of all feature speech and its weights as interest vector.The specific implementation process of this step is consistent with abovementioned steps S401.
In step S502, server 100 utilizes the interest in its scene analysis unit 102 to move towards prediction module 1023, according to interest vector the trend of user interest is predicted, generate a scene information that comprises the user interest trend, and will write among the Cookie of webpage behind this scene information coding.Among the present invention, the scene information that comprises the user interest trend can have various ways, and its generating mode also has multiple.
In generating an exemplary scenario of scene information, interest vector is weighted average computing, the weight the closer to nearest interest vector is obtained big more, also, weight value and its time are inversely proportional in proper order; Then the result after the weighted mean is carried out regularization and handle (making the weight sum is 1), income value is scene information of the present invention.Because in this generation method, it is simple average that interest vector is had more than, but is weighted on average, the value of weight has been considered the drift of scene information, therefore can embody the user interest trend.The scene information that but utilizes this mode to generate remains an integrated value, is partial to remarkable inadequately.
In another exemplary scenario that generates scene information, then utilize the multidimensional time series to carry out computing.Specifically comprise: with described interest vector as the value of multidimensional time series on corresponding time point; Adopt data digging method to predict the next value of this multidimensional seasonal effect in time series then, and should next one value as scene information.In this exemplary scenario, be used to predict that the data digging method of the next value of multidimensional seasonal effect in time series can be multiple, (the Autoregressive IntegratedMoving Average of the autoregression moving average in the statistics for example, ARIMA), and any Forecasting Methodology in the machine learning (for example, neural network, decision tree, support vector machine etc.).
In step S503, select advertisement pointedly according to the scene information behind the coding, and selected advertisement pushing is given the client at user place.Owing to include the factor of user interest trend in this scene information, therefore the selected advertisement nearest demand of can being close to the users is carried out then and is thrown in operation.
In step S504, effect analysis unit 104 is according to advertisement pushing data and each client (client 200, the client 300 of record in the advertisement pushing unit 103 ... client N) Fan Kui user click data calculates the exposure rate and the clicking rate of advertisement.
Among the present invention, the computing method of exposure rate and hit rate can have multiple.In an exemplary scenario, the formula of effect analysis unit 104 calculation exposure rates is as follows: exposure rate=covering number of users/total number of users; Its formula that calculates hit rate is as follows: hit rate=clicks/impression.
In an embodiment of above-mentioned exemplary scenario, gained exposure rate and hit rate such as following table:
| The attribute classification | Number of users | Advertised name | Exposure frequency | Exposure rate | Number of clicks | Hit rate |
| The automobile group | 1,000,000 | BMW S series | 900,000 | 90% 300,000 | 33.3％ |
| Women user | 5,000,000 | LUX beauty soap | 3,000,000 | 60% 2,400,000 | 80％ |
In another embodiment of above-mentioned exemplary scenario, gained exposure rate and hit rate such as following table:
| The attribute classification | Number of users | Advertised name | Exposure frequency | Exposure rate | Number of clicks | Hit rate |
| 25～30 years old the male sex's white collar in Shenzhen | 1,000,000 | The sale advertising of sub-district, South Mountain | 400,000 | 40% 100,000 | 25％ | |
| Family income surpasses 500,000 the women more than 30 years old | 500,000 | The new product clothes | 400,000 | 80% 300,000 | 75％ | |
Certainly, effect analysis unit 104 also can calculate the exposure rate and the hit rate of advertisement by other modes among the present invention, is not limited to above-described method.
At last result of calculation is sent in the scene analysis unit 102, changes step S501, thereby utilize exposure rate and hit rate that the scene analysis process is further optimized.
The above only is preferred embodiment of the present invention, not in order to restriction the present invention, all any modifications of being done within the spirit and principles in the present invention, is equal to and replaces and improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1, a kind of system that pushes advertisement based on scene information, comprise the server and client side, described server comprises a database that is used to store the webpage that the user visits, a webpage that is used for visiting according to the user obtains the scene analysis unit of scene information, with an advertisement pushing unit according to the scene information advertisement delivery, it is characterized in that, comprise in the described scene analysis unit that an interest moves towards prediction module, the webpage predictive user interest that is used for visiting according to the user is moved towards, and generates a scene information that comprises described user interest trend.
2, the system based on scene information propelling movement advertisement according to claim 1 is characterized in that described scene analysis unit also comprises feature speech extraction module, interest vector generation module;
The webpage that described feature speech extraction module is used for browsing from the user extracts the feature speech;
Described interest vector generation module links to each other with feature speech extraction module, is used for calculating the weights of described feature speech at each webpage that the user browsed, and interest vector is formed in described feature speech and weights thereof, and send into interest and move towards prediction module.
3, the system that pushes advertisement based on scene information according to claim 2, it is characterized in that, described server further comprises an effect analysis unit, link to each other with described advertisement pushing unit, be used for user click data, calculate the exposure rate and the clicking rate of advertisement according to advertisement pushing data and client feedback.
4, a kind of equipment that pushes advertisement based on scene information, the i.e. server that links to each other with client, described server comprises that is used to collect the also database of storaging user data, a scene analysis unit that is used for obtaining scene information according to user data, with an advertisement pushing unit according to the scene information advertisement delivery, it is characterized in that, comprise in the described scene analysis unit that an interest moves towards prediction module, the webpage predictive user interest that is used for visiting according to the user is moved towards, and generates a scene information that comprises described user interest trend.
5, the equipment based on scene information propelling movement advertisement according to claim 4 is characterized in that described scene analysis unit also comprises feature speech extraction module, interest vector generation module;
The webpage that described feature speech extraction module is used for browsing from the user extracts the feature speech;
Described interest vector generation module links to each other with feature speech extraction module, is used for calculating the weights of described feature speech at each webpage that the user browsed, and interest vector is formed in described feature speech and weights thereof, and send into interest and move towards prediction module.
6, a kind of method based on scene information propelling movement advertisement is characterized in that, said method comprising the steps of:
A. statistic of user accessing web page, and generate interest vector according to the feature speech that from described webpage, extracts;
B. move towards according to described interest vector predictive user interest, and generate a scene information that comprises the user interest trend;
C. select advertisement according to scene information, and with described advertisement pushing to the client at user place.
7, the method based on scene information propelling movement advertisement according to claim 6 is characterized in that, also comprises before the described steps A: utilize participle technique that webpage is classified.
8, the method based on scene information propelling movement advertisement according to claim 6 is characterized in that described steps A further comprises:
A1. from user accessing web page, extract the feature speech, and with the union of described feature speech as the candidate feature speech;
A2. weights of each in the calculated candidate feature speech, and feature speech and weights thereof be organized into an interest vector.
9, the method based on scene information propelling movement advertisement according to claim 8 is characterized in that described step B comprises:
B11. each component in the described interest vector is weighted on average, wherein the weight value of each component and its time are inversely proportional in proper order;
B12. the weighted mean value to step B11 output carries out the regularization processing, and operation result is encoded as scene information.
10, the method based on scene information propelling movement advertisement according to claim 8 is characterized in that described step B comprises:
B21. with described interest vector as the value of multidimensional time series on corresponding time point;
B22. predict the next value of described multidimensional seasonal effect in time series, and described next value is encoded as scene information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007100761267A CN101079063A (en) | 2007-06-25 | 2007-06-25 | Method, system and apparatus for transmitting advertisement based on scene information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007100761267A CN101079063A (en) | 2007-06-25 | 2007-06-25 | Method, system and apparatus for transmitting advertisement based on scene information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101079063A true CN101079063A (en) | 2007-11-28 |
Family
ID=38906542
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2007100761267A Pending CN101079063A (en) | 2007-06-25 | 2007-06-25 | Method, system and apparatus for transmitting advertisement based on scene information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101079063A (en) |
Cited By (49)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009137978A1 (en) * | 2008-05-14 | 2009-11-19 | 华为技术有限公司 | Method, system and device for presenting advertisement |
| CN101866341A (en) * | 2009-04-17 | 2010-10-20 | 华为技术有限公司 | Information push method, device and system |
| CN101877711A (en) * | 2009-04-28 | 2010-11-03 | 华为技术有限公司 | Social network establishment method and device, and community discovery method and device |
| CN101917664A (en) * | 2010-08-10 | 2010-12-15 | 华为终端有限公司 | Information providing method and device and mobile terminal |
| CN102065145A (en) * | 2010-12-31 | 2011-05-18 | 华为技术有限公司 | Information issuing method, device and system |
| CN102314450A (en) * | 2010-06-30 | 2012-01-11 | 国际商业机器公司 | Be used to strengthen the method and apparatus of web page browsing |
| US20120173341A1 (en) * | 2010-12-31 | 2012-07-05 | Kun Li | Information publishing method, apparatus and system |
| CN102629273A (en) * | 2012-03-19 | 2012-08-08 | 奇智软件（北京）有限公司 | Advertising information pushing method based on browser, device and browser system |
| CN102663616A (en) * | 2012-03-19 | 2012-09-12 | 北京国双科技有限公司 | Method and system for measuring web advertising effectiveness based on multiple-contact attribution model |
| CN102685215A (en) * | 2012-04-18 | 2012-09-19 | 华为技术有限公司 | Method, device and system for online monitoring of mobile terminal |
| CN102710655A (en) * | 2012-06-15 | 2012-10-03 | 深圳市方果网络科技有限公司 | Network interaction system and method based on sports activities |
| CN102722832A (en) * | 2012-01-05 | 2012-10-10 | 合一网络技术(北京)有限公司 | Online video advertisement refinement targeting delivery method |
| CN102902791A (en) * | 2012-09-29 | 2013-01-30 | 北京奇虎科技有限公司 | Webpage sorted storage system and method |
| CN103067470A (en) * | 2012-12-21 | 2013-04-24 | 北京奇虎科技有限公司 | Method, server and system for pushing information to browser |
| CN103166930A (en) * | 2011-12-15 | 2013-06-19 | 腾讯科技（深圳）有限公司 | Method and system for pushing network information |
| CN103177384A (en) * | 2013-04-16 | 2013-06-26 | 高鹏 | Network advertisement putting method based on user interest spectrum |
| CN103229169A (en) * | 2010-11-25 | 2013-07-31 | 三星电子株式会社 | Content-providing method and system |
| CN103235824A (en) * | 2013-05-06 | 2013-08-07 | 上海河广信息科技有限公司 | Method and system for determining web page texts users interested in according to browsed web pages |
| CN103368898A (en) * | 2012-03-26 | 2013-10-23 | 中兴通讯股份有限公司 | Method and system for accomplishing information push |
| CN103577478A (en) * | 2012-08-06 | 2014-02-12 | 腾讯科技（深圳）有限公司 | Web page pushing method and system |
| CN103577504A (en) * | 2012-08-10 | 2014-02-12 | 华为技术有限公司 | Method and device for putting personalized contents |
| WO2014040560A1 (en) * | 2012-09-17 | 2014-03-20 | 腾讯科技（深圳）有限公司 | Method and device for pushing media information in real time |
| CN103714063A (en) * | 2012-09-28 | 2014-04-09 | 国际商业机器公司 | Data analysis method and data analysis system |
| CN103793403A (en) * | 2012-10-31 | 2014-05-14 | 中国电信股份有限公司 | Method and device for forwarding and searching information related with results |
| CN103870671A (en) * | 2012-12-18 | 2014-06-18 | 北京思博途信息技术有限公司 | Method and device for extracting user sample from Cookies |
| CN103885951A (en) * | 2012-12-19 | 2014-06-25 | 阿里巴巴集团控股有限公司 | Graphics and text information releasing and generating method and graphics and text information releasing and generating device |
| CN103888498A (en) * | 2012-12-21 | 2014-06-25 | 腾讯科技（深圳）有限公司 | Information pushing method and apparatus, terminal and server |
| CN104065672A (en) * | 2013-03-18 | 2014-09-24 | 华为技术有限公司 | Advertisement pushing method, client and advertisement pushing system |
| CN104217353A (en) * | 2014-08-20 | 2014-12-17 | 苏州海云网络科技有限公司 | Webpage advertisement directional pushing system |
| CN104469257A (en) * | 2013-09-24 | 2015-03-25 | 谷歌公司 | System and method for shoppable video feed |
| CN104462593A (en) * | 2014-12-29 | 2015-03-25 | 北京奇虎科技有限公司 | Method and device for providing user personalized resource message pushing |
| CN105046531A (en) * | 2015-08-07 | 2015-11-11 | 百度在线网络技术（北京）有限公司 | Online delivery information display method and apparatus |
| CN105302807A (en) * | 2014-06-06 | 2016-02-03 | 腾讯科技（深圳）有限公司 | Method and apparatus for obtaining information category |
| CN102033883B (en) * | 2009-09-29 | 2016-03-02 | 阿里巴巴集团控股有限公司 | A kind of method, Apparatus and system improving data transmission speed of website |
| CN105589866A (en) * | 2014-10-22 | 2016-05-18 | 联想(北京)有限公司 | Information display method and apparatus |
| CN106161627A (en) * | 2016-07-12 | 2016-11-23 | 百度在线网络技术（北京）有限公司 | Method and device for pushed information |
| CN106331076A (en) * | 2016-08-17 | 2017-01-11 | 北京小米移动软件有限公司 | Advertisement push method and apparatus |
| CN106651681A (en) * | 2017-01-12 | 2017-05-10 | 安徽大尺度网络传媒有限公司 | Real estate information service system based on user data analysis |
| CN106921783A (en) * | 2015-12-28 | 2017-07-04 | 卡桑德电子科技（扬州）有限公司 | Use terminal device, the system and method for communication and social networking application when driving safely |
| WO2017128358A1 (en) * | 2016-01-30 | 2017-08-03 | 盛玉伟 | Method and system for new house opening notification on real estate network |
| CN107483554A (en) * | 2017-07-25 | 2017-12-15 | 中天宽带技术有限公司 | Network traffics based on ONU carry out the supplying system and method for machine learning targeted ads |
| CN107590244A (en) * | 2017-09-14 | 2018-01-16 | 深圳市和讯华谷信息技术有限公司 | The recognition methods of mobile device Below-the-line scene and device |
| CN108133294A (en) * | 2018-01-10 | 2018-06-08 | 阳光财产保险股份有限公司 | Forecasting Methodology and device based on information sharing |
| CN108170673A (en) * | 2017-12-26 | 2018-06-15 | 北京百度网讯科技有限公司 | The recognition methods of information style and device based on artificial intelligence |
| CN108459936A (en) * | 2017-02-20 | 2018-08-28 | 北京畅游时空软件技术有限公司 | A kind of accurate statistical method and device based on contents pattern blocked |
| CN105243087B (en) * | 2015-09-09 | 2018-08-31 | 浙江网新恒天软件有限公司 | IT syndication Personality of readingization recommends method |
| US10475047B2 (en) | 2009-08-28 | 2019-11-12 | Resonate Networks, Inc. | Method and apparatus for delivering targeted content to website visitors |
| CN110580631A (en) * | 2018-06-07 | 2019-12-17 | 北京奇虎科技有限公司 | advertisement putting method and device |
| CN112258207A (en) * | 2020-06-16 | 2021-01-22 | 北京沃东天骏信息技术有限公司 | Advertisement flow determination method, device, equipment and storage medium |
-
2007
- 2007-06-25 CN CNA2007100761267A patent/CN101079063A/en active Pending
Cited By (75)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101582911B (en) * | 2008-05-14 | 2014-12-03 | 华为技术有限公司 | Method, system and device for presenting advertisement |
| WO2009137978A1 (en) * | 2008-05-14 | 2009-11-19 | 华为技术有限公司 | Method, system and device for presenting advertisement |
| CN101866341A (en) * | 2009-04-17 | 2010-10-20 | 华为技术有限公司 | Information push method, device and system |
| CN101877711A (en) * | 2009-04-28 | 2010-11-03 | 华为技术有限公司 | Social network establishment method and device, and community discovery method and device |
| CN101877711B (en) * | 2009-04-28 | 2013-08-28 | 华为技术有限公司 | Social network establishment method and device, and community discovery method and device |
| US10475047B2 (en) | 2009-08-28 | 2019-11-12 | Resonate Networks, Inc. | Method and apparatus for delivering targeted content to website visitors |
| CN102033883B (en) * | 2009-09-29 | 2016-03-02 | 阿里巴巴集团控股有限公司 | A kind of method, Apparatus and system improving data transmission speed of website |
| US8577900B2 (en) | 2010-06-30 | 2013-11-05 | International Business Machines Corporation | Method and apparatus for enhancing webpage browsing |
| CN102314450A (en) * | 2010-06-30 | 2012-01-11 | 国际商业机器公司 | Be used to strengthen the method and apparatus of web page browsing |
| CN101917664A (en) * | 2010-08-10 | 2010-12-15 | 华为终端有限公司 | Information providing method and device and mobile terminal |
| CN101917664B (en) * | 2010-08-10 | 2014-12-24 | 华为终端有限公司 | Information providing method and device and mobile terminal |
| CN103229169B (en) * | 2010-11-25 | 2016-06-01 | 三星电子株式会社 | Content providing and system |
| US9465863B2 (en) | 2010-11-25 | 2016-10-11 | Samsung Electronics Co., Ltd. | Content-providing method and system |
| CN103229169A (en) * | 2010-11-25 | 2013-07-31 | 三星电子株式会社 | Content-providing method and system |
| CN102065145B (en) * | 2010-12-31 | 2013-04-17 | 华为技术有限公司 | Information issuing method, device and system |
| US20120173341A1 (en) * | 2010-12-31 | 2012-07-05 | Kun Li | Information publishing method, apparatus and system |
| CN102065145A (en) * | 2010-12-31 | 2011-05-18 | 华为技术有限公司 | Information issuing method, device and system |
| CN103166930B (en) * | 2011-12-15 | 2016-04-13 | 腾讯科技（深圳）有限公司 | The method and system of pushing network information |
| CN103166930A (en) * | 2011-12-15 | 2013-06-19 | 腾讯科技（深圳）有限公司 | Method and system for pushing network information |
| CN102722832A (en) * | 2012-01-05 | 2012-10-10 | 合一网络技术(北京)有限公司 | Online video advertisement refinement targeting delivery method |
| CN102722832B (en) * | 2012-01-05 | 2015-08-19 | 合一网络技术(北京)有限公司 | Online Video advertisement becomes more meticulous directed put-on method |
| CN102663616A (en) * | 2012-03-19 | 2012-09-12 | 北京国双科技有限公司 | Method and system for measuring web advertising effectiveness based on multiple-contact attribution model |
| CN102629273A (en) * | 2012-03-19 | 2012-08-08 | 奇智软件（北京）有限公司 | Advertising information pushing method based on browser, device and browser system |
| CN102629273B (en) * | 2012-03-19 | 2014-11-05 | 北京奇虎科技有限公司 | Advertising information pushing method based on browser, device and browser system |
| CN103368898A (en) * | 2012-03-26 | 2013-10-23 | 中兴通讯股份有限公司 | Method and system for accomplishing information push |
| CN102685215A (en) * | 2012-04-18 | 2012-09-19 | 华为技术有限公司 | Method, device and system for online monitoring of mobile terminal |
| CN102710655B (en) * | 2012-06-15 | 2015-12-16 | 深圳市方果网络科技有限公司 | A kind of network interaction system and method based on sports |
| CN102710655A (en) * | 2012-06-15 | 2012-10-03 | 深圳市方果网络科技有限公司 | Network interaction system and method based on sports activities |
| CN103577478B (en) * | 2012-08-06 | 2015-07-29 | 腾讯科技（深圳）有限公司 | Web page push method and system |
| CN103577478A (en) * | 2012-08-06 | 2014-02-12 | 腾讯科技（深圳）有限公司 | Web page pushing method and system |
| WO2014023121A1 (en) * | 2012-08-10 | 2014-02-13 | 华为技术有限公司 | Method and device for launching individual content |
| CN103577504A (en) * | 2012-08-10 | 2014-02-12 | 华为技术有限公司 | Method and device for putting personalized contents |
| WO2014040560A1 (en) * | 2012-09-17 | 2014-03-20 | 腾讯科技（深圳）有限公司 | Method and device for pushing media information in real time |
| US11222375B2 (en) | 2012-09-28 | 2022-01-11 | International Business Machines Corporation | Data analysis method and system thereof |
| CN103714063B (en) * | 2012-09-28 | 2017-08-04 | 国际商业机器公司 | Data analysing method and its system |
| US11176586B2 (en) | 2012-09-28 | 2021-11-16 | International Business Machines Corporation | Data analysis method and system thereof |
| CN103714063A (en) * | 2012-09-28 | 2014-04-09 | 国际商业机器公司 | Data analysis method and data analysis system |
| CN102902791A (en) * | 2012-09-29 | 2013-01-30 | 北京奇虎科技有限公司 | Webpage sorted storage system and method |
| CN102902791B (en) * | 2012-09-29 | 2016-08-03 | 北京奇虎科技有限公司 | Web page classification storage system and method |
| CN103793403B (en) * | 2012-10-31 | 2017-06-23 | 中国电信股份有限公司 | Push the method and apparatus with Search Results associated information |
| CN103793403A (en) * | 2012-10-31 | 2014-05-14 | 中国电信股份有限公司 | Method and device for forwarding and searching information related with results |
| CN103870671A (en) * | 2012-12-18 | 2014-06-18 | 北京思博途信息技术有限公司 | Method and device for extracting user sample from Cookies |
| CN103885951A (en) * | 2012-12-19 | 2014-06-25 | 阿里巴巴集团控股有限公司 | Graphics and text information releasing and generating method and graphics and text information releasing and generating device |
| CN103888498B (en) * | 2012-12-21 | 2018-04-13 | 腾讯科技（深圳）有限公司 | Information-pushing method, device, terminal and server |
| CN103888498A (en) * | 2012-12-21 | 2014-06-25 | 腾讯科技（深圳）有限公司 | Information pushing method and apparatus, terminal and server |
| CN103067470B (en) * | 2012-12-21 | 2016-08-03 | 北京奇虎科技有限公司 | A kind of method to browser pushed information, server and system |
| CN103067470A (en) * | 2012-12-21 | 2013-04-24 | 北京奇虎科技有限公司 | Method, server and system for pushing information to browser |
| CN104065672B (en) * | 2013-03-18 | 2018-03-06 | 华为技术有限公司 | Advertisement sending method, client and advertisement push system |
| CN104065672A (en) * | 2013-03-18 | 2014-09-24 | 华为技术有限公司 | Advertisement pushing method, client and advertisement pushing system |
| CN103177384A (en) * | 2013-04-16 | 2013-06-26 | 高鹏 | Network advertisement putting method based on user interest spectrum |
| CN103235824A (en) * | 2013-05-06 | 2013-08-07 | 上海河广信息科技有限公司 | Method and system for determining web page texts users interested in according to browsed web pages |
| CN104469257A (en) * | 2013-09-24 | 2015-03-25 | 谷歌公司 | System and method for shoppable video feed |
| US10346496B2 (en) | 2014-06-06 | 2019-07-09 | Tencent Technology (Shenzhen) Company Limited | Information category obtaining method and apparatus |
| CN105302807A (en) * | 2014-06-06 | 2016-02-03 | 腾讯科技（深圳）有限公司 | Method and apparatus for obtaining information category |
| CN104217353A (en) * | 2014-08-20 | 2014-12-17 | 苏州海云网络科技有限公司 | Webpage advertisement directional pushing system |
| CN105589866A (en) * | 2014-10-22 | 2016-05-18 | 联想(北京)有限公司 | Information display method and apparatus |
| CN104462593A (en) * | 2014-12-29 | 2015-03-25 | 北京奇虎科技有限公司 | Method and device for providing user personalized resource message pushing |
| CN105046531A (en) * | 2015-08-07 | 2015-11-11 | 百度在线网络技术（北京）有限公司 | Online delivery information display method and apparatus |
| CN105243087B (en) * | 2015-09-09 | 2018-08-31 | 浙江网新恒天软件有限公司 | IT syndication Personality of readingization recommends method |
| CN106921783A (en) * | 2015-12-28 | 2017-07-04 | 卡桑德电子科技（扬州）有限公司 | Use terminal device, the system and method for communication and social networking application when driving safely |
| WO2017128358A1 (en) * | 2016-01-30 | 2017-08-03 | 盛玉伟 | Method and system for new house opening notification on real estate network |
| CN106161627A (en) * | 2016-07-12 | 2016-11-23 | 百度在线网络技术（北京）有限公司 | Method and device for pushed information |
| CN106331076A (en) * | 2016-08-17 | 2017-01-11 | 北京小米移动软件有限公司 | Advertisement push method and apparatus |
| CN106651681A (en) * | 2017-01-12 | 2017-05-10 | 安徽大尺度网络传媒有限公司 | Real estate information service system based on user data analysis |
| CN108459936A (en) * | 2017-02-20 | 2018-08-28 | 北京畅游时空软件技术有限公司 | A kind of accurate statistical method and device based on contents pattern blocked |
| CN108459936B (en) * | 2017-02-20 | 2021-05-14 | 北京畅游时空软件技术有限公司 | Accurate statistical method and device based on content modularization |
| CN107483554A (en) * | 2017-07-25 | 2017-12-15 | 中天宽带技术有限公司 | Network traffics based on ONU carry out the supplying system and method for machine learning targeted ads |
| CN107590244A (en) * | 2017-09-14 | 2018-01-16 | 深圳市和讯华谷信息技术有限公司 | The recognition methods of mobile device Below-the-line scene and device |
| CN107590244B (en) * | 2017-09-14 | 2020-04-17 | 深圳市和讯华谷信息技术有限公司 | Method and device for identifying offline activity scene of mobile equipment |
| CN108170673A (en) * | 2017-12-26 | 2018-06-15 | 北京百度网讯科技有限公司 | The recognition methods of information style and device based on artificial intelligence |
| CN108170673B (en) * | 2017-12-26 | 2021-08-24 | 北京百度网讯科技有限公司 | Information tone identification method and device based on artificial intelligence |
| CN108133294B (en) * | 2018-01-10 | 2020-12-04 | 阳光财产保险股份有限公司 | Prediction method and device based on information sharing |
| CN108133294A (en) * | 2018-01-10 | 2018-06-08 | 阳光财产保险股份有限公司 | Forecasting Methodology and device based on information sharing |
| CN110580631A (en) * | 2018-06-07 | 2019-12-17 | 北京奇虎科技有限公司 | advertisement putting method and device |
| CN112258207A (en) * | 2020-06-16 | 2021-01-22 | 北京沃东天骏信息技术有限公司 | Advertisement flow determination method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101079063A (en) | Method, system and apparatus for transmitting advertisement based on scene information | |
| US9767182B1 (en) | Classification of search queries | |
| US7870474B2 (en) | System and method for smoothing hierarchical data using isotonic regression | |
| US7519588B2 (en) | Keyword characterization and application | |
| CN101593200B (en) | Method for classifying Chinese webpages based on keyword frequency analysis | |
| CN1186737C (en) | Method and system for queuing uncalled web based on path | |
| CN101911042B (en) | The relevance ranking of the browser history of user | |
| US9798820B1 (en) | Classification of keywords | |
| US8311882B2 (en) | System and method for forecasting an inventory of online advertisement impressions for targeting impression attributes | |
| US8782037B1 (en) | System and method for mark-up language document rank analysis | |
| CN100483408C (en) | Method and apparatus for establishing link structure between multiple documents | |
| CN103514191A (en) | Method and device for determining keyword matching mode of target popularization information | |
| CN112508609B (en) | Crowd expansion prediction method, device, equipment and storage medium | |
| CN101192235A (en) | Method, system and equipment for delivering advertisement based on user feature | |
| CN1764916A (en) | Method and apparatus for frequency count | |
| US9171045B2 (en) | Recommending queries according to mapping of query communities | |
| CN101814083A (en) | Automatic webpage classification method and system | |
| CN103631803A (en) | Method, device and server for advertisement orientation based on input behaviors | |
| CN101694658A (en) | Method for constructing webpage crawler based on repeated removal of news | |
| CN102426610A (en) | Microblog rank searching method and microblog searching engine | |
| CN101685521A (en) | Method for showing advertisements in webpage and system | |
| CN103914468A (en) | Method and device for searching for released information | |
| WO2016107354A1 (en) | Method and apparatus for providing user personalised resource message pushing | |
| CN111178950A (en) | User portrait construction method and device and computing equipment | |
| CN102236710A (en) | Method and equipment for displaying news information in query result |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |