CN105224617A - Recommendation method, device and system for broadcast content - Google Patents
Recommendation method, device and system for broadcast content Download PDFInfo
- Publication number
- CN105224617A CN105224617A CN201510595693.8A CN201510595693A CN105224617A CN 105224617 A CN105224617 A CN 105224617A CN 201510595693 A CN201510595693 A CN 201510595693A CN 105224617 A CN105224617 A CN 105224617A
- Authority
- CN
- China
- Prior art keywords
- fixed reference
- reference feature
- user
- group
- content
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a method, a device and a system for recommending broadcast content. After user information and content information of candidate broadcast content are obtained, a first group of externally stored reference features and a second group of internally stored reference features related to the user information are obtained; searching for a reference feature matched with the content information in the candidate broadcast content from the first group of reference features and the second group of reference features related to the user information; and after the reference characteristics matched with the content information of the candidate broadcast contents are found, screening out at least one broadcast content from the candidate broadcast contents according to the found reference characteristics to recommend the broadcast content to the user. By adopting the embodiment of the invention, the time delay query can be reduced, and the high-frequency query can be ensured.
Description
Technical field
The present invention relates to Internet technical field, particularly relate to a kind of recommend method of broadcasted content, Apparatus and system.
Background technology
Current, server can push multiple broadcasted content to user terminal, as advertisement, and news etc.For advertisement, in the process that the ad click rate of advertisement commending system is estimated, by clicking with the ratio of expected click, (English: ClickOverExpectedClick, abbreviation: COEC) feature, estimates the clicking rate of multiple advertisement to be recommended.Concrete, COEC feature can comprise characteristic key words and eigenwert, and characteristic key words can be divided into user property and advertisement attributes.The eigenwert that different characteristic key words is corresponding different, therefore, according to the information searching of active user and candidate locations to characteristic of correspondence keyword, characteristic key words characteristic of correspondence value can be obtained, thus eigenwert more much higher candidate locations can be filtered out in advertisement commending system.
But the combination of the content information of the broadcasted content of current user information and some candidates may correspond to the characteristic key words of multiple COEC feature, and in the recommendation of a broadcasted content, usually need to estimate the candidate broadcast content of hundreds of or more quantity.Due to the substantial amounts of candidate's broadcasted content characteristics of correspondence such as COEC, and feature renewal frequency is high, usually by characteristic storage in external memory storage.Therefore, in the process of the characteristic key words of the content information query characteristics according to current user information and candidate broadcast content, need externally storer to send inquiry request, and inquiry times is huge, cause inquiry time delay, cause the operational efficiency of the commending system of broadcasted content lower.
Summary of the invention
Embodiments provide a kind of recommend method of broadcasted content, Apparatus and system.Adopt the embodiment of the present invention, time delay inquiry can be reduced, and can high frequency queries be ensured.
The embodiment of the present invention provides a kind of recommend method of broadcasted content in first aspect, and the method can comprise:
When after the content information getting user profile and candidate broadcast content, obtain first group of fixed reference feature of the exterior storage relevant to described user profile and second group of fixed reference feature of storage inside;
In the described first group fixed reference feature relevant to described user profile got and described second group of fixed reference feature, search the fixed reference feature matched with the content information of described candidate broadcast content;
After finding the fixed reference feature matched with the content information of described candidate broadcast content, according to the described fixed reference feature found, from described candidate broadcast content, filter out at least one broadcasted content to recommend user.
As optional embodiment, described fixed reference feature comprises characteristic key words and eigenwert; Wherein,
Described characteristic key words comprises user property and broadcasted content attribute, and wherein, described user property comprises user ID;
Described eigenwert at least comprises number of clicks and expected click number of times.
As optional embodiment, first group of fixed reference feature of the exterior storage that described acquisition is relevant to described user profile and second group of fixed reference feature of storage inside, comprising:
Externally storer sends inquiry request, and wherein, described inquiry request comprises the user ID in described user profile;
Obtain first group of fixed reference feature of described external memory storage feedback, wherein, the user property in described first group of fixed reference feature comprises described user ID;
From internal storage, obtain second group of fixed reference feature, wherein, the user property in described second group of fixed reference feature comprises the user profile except described user ID.
As optional embodiment, in described described first group of fixed reference feature relevant to described user profile getting and described second group of fixed reference feature, searching the fixed reference feature matched with the content information of described candidate broadcast content, comprising:
The content information of described user profile and described candidate broadcast content is carried out combining to form combined information;
The fixed reference feature matched with described combined information is searched in described first group of fixed reference feature and described second group of fixed reference feature.
As optional embodiment, the fixed reference feature found described in described basis, filters out at least one broadcasted content to recommend user, comprising from described candidate broadcast content:
The eigenwert of the fixed reference feature found described in acquisition;
According to described eigenwert, described fixed reference feature is sorted;
Choose at least one fixed reference feature in described sequence;
Broadcasted content corresponding for broadcasted content attribute in described fixed reference feature is recommended user.
As optional embodiment, described according to described eigenwert, described fixed reference feature is sorted, comprising:
According to the ratio of the described number of clicks in described eigenwert and described expected click number of times, what calculate described fixed reference feature estimates clicking rate;
Estimate clicking rate order from big to small according to described, described fixed reference feature is sorted.
The second aspect of the embodiment of the present invention provides a kind of recommendation apparatus of broadcasted content, and this device can comprise:
Acquisition module, for when after the content information getting user profile and candidate broadcast content, obtains first group of fixed reference feature of the exterior storage relevant to described user profile and second group of fixed reference feature of storage inside;
Search module, in the described first group fixed reference feature relevant to described user profile got and described second group of fixed reference feature, search the fixed reference feature matched with the content information of described candidate broadcast content;
Screening module, for after finding the fixed reference feature matched with the advertising message of described candidate locations, according to the described fixed reference feature found, filters out at least one broadcasted content to recommend user from described candidate broadcast content.
As optional embodiment, described fixed reference feature comprises characteristic key words and eigenwert; Wherein,
Described characteristic key words comprises user property and broadcasted content attribute, and wherein, described user property comprises user ID;
Described eigenwert at least comprises number of clicks and expected click number of times.
As optional embodiment, described acquisition module comprises:
Transmitting element, send inquiry request for externally storer, wherein, described inquiry request comprises the user ID in described user profile;
First acquiring unit, for obtaining first group of fixed reference feature of described external memory storage feedback, wherein, the user property in described first group of fixed reference feature comprises described user ID;
Second acquisition unit, for obtaining second group of fixed reference feature from internal storage, wherein, the user property in described second group of fixed reference feature comprises the user profile except described user ID.
As optional embodiment, described in search module and comprise:
Assembled unit, for being undertaken combining to form combined information by the content information of described user profile and described candidate broadcast content;
Search unit, for searching the fixed reference feature matched with described combined information in described first group of fixed reference feature and described second group of fixed reference feature.
As optional embodiment, described screening module comprises:
3rd acquiring unit, for the eigenwert of fixed reference feature found described in obtaining;
Sequencing unit, for according to described eigenwert, sorts described fixed reference feature;
Choose unit, for choosing at least one fixed reference feature in described sequence;
Recommendation unit, for recommending user by broadcasted content corresponding for broadcasted content attribute in described fixed reference feature.
As optional embodiment, described sequencing unit comprises:
Computation subunit, for the ratio according to the described number of clicks in described eigenwert and described expected click number of times, what calculate described fixed reference feature estimates clicking rate;
Sequence subelement, for estimating clicking rate order from big to small according to described, sorts to described fixed reference feature.
The embodiment of the present invention third aspect provides a kind of commending system of broadcasted content.This system can comprise estimates server and external memory storage;
Wherein, described external memory storage is for storing the fixed reference feature relevant to user ID;
Described server of estimating comprises internal storage and processor;
Described internal storage identifies relevant fixed reference feature for storing to non-user;
Described processor for call in described external memory storage store described in the fixed reference feature relevant to user ID, and call in described internal storage the described and non-user stored and identify relevant fixed reference feature, and according to the described fixed reference feature called, filter out at least one broadcasted content to recommend user.
In the embodiment of the present invention, when after the content information getting user profile and candidate broadcast content, first group of fixed reference feature of the exterior storage relevant to described user profile and second group of fixed reference feature of storage inside can be obtained respectively, and the fixed reference feature by getting, search the fixed reference feature matched with the content information of user profile and broadcasted content, after finding corresponding fixed reference feature, can screen candidate broadcast content according to fixed reference feature, to filter out the broadcasted content high with user's degree of correlation.By being pre-stored in external memory storage and internal storage with reference to feature, can realizing once calling and user-dependent fixed reference feature from external memory storage, reducing the inquiry times to external memory storage, thus can inquiry time delay be reduced.
Accompanying drawing explanation
In order to be illustrated more clearly in the technical scheme of the embodiment of the present invention, below the accompanying drawing used required in describing embodiment is briefly described, apparently, accompanying drawing in the following describes is some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.
Fig. 1 is the process flow diagram of an embodiment of the recommend method of a kind of broadcasted content in the present invention;
Fig. 2 is the process flow diagram of another embodiment of the recommend method of a kind of broadcasted content in the present invention;
Fig. 3 is the structural representation of an embodiment of the recommendation apparatus of a kind of broadcasted content in the present invention;
Fig. 4 is the structural representation of another embodiment of the recommendation apparatus of a kind of broadcasted content in the present invention;
Fig. 5 is the structural representation of the commending system of a kind of broadcasted content in the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
Below with reference to the accompanying drawings embodiments of the invention are described.Wherein, the device described in the embodiment of the present invention can comprise server etc. can call data in inside or external memory storage, and can carry out device or the equipment of data processing, and described in the embodiment of the present invention, device is described for server.
See Fig. 1, it is the process flow diagram of an embodiment of the recommend method of a kind of broadcasted content in the present invention.Wherein, method described in the embodiment of the present invention can be realized by device correspondence.The method can comprise the following steps.
Step S101, when after the content information getting user profile and candidate broadcast content, obtains first group of fixed reference feature of the exterior storage relevant to described user profile and second group of fixed reference feature of storage inside.
In one embodiment, after the broadcasted content in server recommends platform to obtain the content information of user profile and candidate broadcast content, can be one or more broadcasted contents in user's recommended candidate broadcasted content according to user profile and content information.Wherein, server obtains user profile etc. by terminal to report.Optionally, user profile can comprise the information such as user ID, age of user, sex, interest, city, user place, country and/or the interested city of user, country; Content information can comprise broadcast content identification, the content source mark of broadcasted content, popularization plan mark, broadcasted content classification, broadcasted content represent the information such as form, broadcasted content source.
After getting and in advance the user profile of recommending broadcast contents and the content information of multiple candidate broadcast content being carried out to it, first group of fixed reference feature of the exterior storage relevant to user profile and second group of fixed reference feature of storage inside can be obtained.Concrete, when server by broadcasted content recommend platform carry out broadcasted content recommend time, the important references feature need recommended as broadcasted content by the fixed reference feature of server internal or outside pre-stored.Optionally, as being user's recommended advertisements, then fixed reference feature can be COEC feature, wherein, as the fixed reference features such as COEC feature can comprise characteristic key words and eigenwert, characteristic key words can comprise user property and broadcasted content attribute, and user property is corresponding with user profile, and broadcasted content attribute is corresponding with content information.Namely by searching the fixed reference feature of outside or inside pre-stored, obtain the eigenwert of the fixed reference feature matched with current user information and content information, then the contrast by eigenwert obtains one or more broadcasted content high with user's degree of correlation.
Optionally, during the fixed reference feature of exterior storage stores by Redis.Wherein, the internal memory that Redis can combine multiple machine forms a memory system, and supports with < characteristic key words, and the mode of eigenwert > stores.Wherein, the fixed reference feature that low frequency is inquired about and memory space is large can be stored in the external memory storage that server connects.The fixed reference feature comprising user ID as user property is stored in external memory storage, when being a certain user's recommending broadcast contents, from external memory storage, only transfers the fixed reference feature that the user ID of this user is relevant; And the fixed reference feature that other non-user mark user property can not being comprised user ID is correlated with is stored in internal storage, supports high frequency inquiry.
Step S102, in the described first group fixed reference feature relevant to described user profile got and described second group of fixed reference feature, searches the fixed reference feature matched with the content information of described candidate broadcast content.
In one embodiment, according to user profile, the first group of fixed reference feature matched with user profile can be obtained from external memory storage, and second group of fixed reference feature can be obtained from internal storage.Optionally, the fixed reference feature obtained can be placed in the buffer, thus can further improve search efficiency.After getting the fixed reference feature relevant to active user, search the fixed reference feature matched with the content information of current candidate broadcasted content.Concrete, if fixed reference feature is COEC feature, then the data layout of feature can be < user property, broadcasted content attribute, number of clicks (c), expected click number of times (ec), COEC>, wherein, number of clicks (c), expected click number of times (ec), COEC is the eigenwert of COEC feature.The COEC feature that the content search that then can comprise according to the content comprised in user profile and some content informations matches.If user ID is user 1, broadcasted content classification 1 is comprised in content information, then can search fixed reference feature is < user 1, broadcasted content classification, the fixed reference feature of broadcasted content classification 1 whether is had in c, ec, COEC>, if exist, then show that this fixed reference feature and content information and user profile match.
Optionally, matched and searched successively can be carried out to candidate broadcast content, as searched the fixed reference feature matched according to the content information of a certain broadcasted content, after finding the fixed reference feature that this broadcasted content matches, can continue to search next broadcasted content according to the same manner.
Step S103, after finding the fixed reference feature matched with the content information of described candidate broadcast content, according to the described fixed reference feature found, filters out at least one broadcasted content to recommend user from described candidate broadcast content.
In one embodiment, after finding the fixed reference feature matched with the content information of candidate broadcast content and user profile, can according to the fixed reference feature found, candidate broadcast content is screened, filters out the one or more broadcasted contents high with active user's degree of correlation to recommend user.Optionally, by iteration decision tree, (English: GradientBoostingDecisionTree is called for short: gdbt) model screens the content information of candidate broadcast content.By the eigenwert of the fixed reference feature of the coupling of acquisition, and Corresponding matching broadcasted content can be obtained according to this eigenwert estimate clicking rate, thus according to the size estimating clicking rate, candidate broadcast content can be sorted, recommend user by estimating the clicking rate high one or more broadcasted contents that sort.Optionally, according to the concrete form in content information, broadcasted content can be promoted to the terminal providing user profile.
In the embodiment of the present invention, when after the content information getting user profile and candidate broadcast content, first group of fixed reference feature of the exterior storage relevant to described user profile and second group of fixed reference feature of storage inside can be obtained respectively, and the fixed reference feature by getting, search the fixed reference feature matched with user profile and content information, after finding corresponding fixed reference feature, can screen candidate broadcast content according to fixed reference feature, to filter out the broadcasted content high with user's degree of correlation.By being pre-stored in external memory storage and internal storage with reference to feature, can realizing once calling and user-dependent fixed reference feature from external memory storage, reducing the inquiry times to external memory storage, thus can inquiry time delay be reduced.
See Fig. 2, it is the process flow diagram of another embodiment of the recommend method of a kind of broadcasted content in the present invention.Wherein, method described in the embodiment of the present invention can be realized by device correspondence.The method can comprise the following steps.
Step S201, when after the content information getting user profile and candidate broadcast content, externally storer sends inquiry request, and wherein, described inquiry request comprises the user ID in described user profile.
In one embodiment, after the broadcasted content in server recommends platform to obtain the content information of user profile and candidate broadcast content, can be one or more broadcasted contents in user's recommended candidate broadcasted content according to user profile and content information.Wherein, server obtains user profile etc. by terminal to report.Optionally, user profile can comprise the information such as user ID, age of user, sex, interest, city, user place, country and/or the interested city of user, country; Content information can comprise broadcast content identification, the content source mark of broadcasted content, popularization plan mark, broadcasted content classification, broadcasted content represent the information such as form, broadcasted content source.Fixed reference feature can be COEC feature, and wherein, COEC feature can comprise characteristic key words and eigenwert, and characteristic key words can comprise user property and broadcasted content attribute, and user property is corresponding with user profile, and broadcasted content attribute is corresponding with content information.
In one embodiment, during the fixed reference feature of exterior storage stores by Redis.Wherein, the internal memory that Redis can combine multiple machine forms a memory system, and supports with < characteristic key words, and the mode of eigenwert > stores.After server gets user profile, user ID can be sent to external memory storage by query statement, namely in Redis storer.Wherein, the fixed reference feature that user ID is relevant is stored in external memory storage.Due to user ID and customer volume linear, therefore in external memory storage, store most of fixed reference feature in fixed reference feature storehouse, inquiry request is sent by externally storer, disposablely can call fixed reference feature corresponding to active user, thus achieve memory space greatly and the effect that reduces of the number of times externally inquired about.
Step S202, obtain first group of fixed reference feature of described external memory storage feedback, wherein, the user property in described first group of fixed reference feature comprises described user ID.
In one embodiment, after external memory storage receives the inquiry request of server, the user ID in inquiry request can be parsed, and the fixed reference feature of this user ID can be comprised in inquiring user attribute, and the fixed reference feature comprising this user ID can be fed back to server with the form of group.After server gets first group of fixed reference feature of external memory storage feedback, namely get the fixed reference feature relevant to user ID.If one of them fixed reference feature in retrievable first group of fixed reference feature is < user ID 1, broadcast content identification k, broadcasted content type m, c, ec, COEC> etc.
Step S203, obtains second group of fixed reference feature from internal storage, and wherein, the user property in described second group of fixed reference feature comprises the user profile except described user ID.
In one embodiment, in user profile, also can comprise other user profile except user ID, as user interest etc., then can obtain the fixed reference feature that the non-user mark of pre-stored is relevant from internal storage.If one of them fixed reference feature in retrievable second group of fixed reference feature is < user interest n, broadcast content identification i, content source mark j, c, ec, the COEC> etc. of broadcasted content.Wherein, internal storage can back-level server high frequency inquiry, as user interest can comprise P, broadcasted content type comprises Q, when needing to search the fixed reference feature all mated user interest and broadcasted content type, need inquiry P*Q time, enquiry frequency is higher, but can realize the low effect of inquiry time delay in internal storage.
Step S204, is undertaken combining to form combined information by the content information of described user profile and described candidate broadcast content.
In one embodiment, after getting the fixed reference feature relevant to user profile, namely by being combined by the content information of user profile and candidate broadcast content, in the first group of fixed reference feature got and second group of fixed reference feature, whether there is the fixed reference feature mated with combined information to search further.Illustrate, data layout due to fixed reference feature is < user property, broadcasted content attribute, eigenwert >, the data layout of combined information can be set to < user profile, content information >.Wherein, if according to the difference of user profile and content information, may be combined with into different combined informations.As combined information can be < user ID 1, broadcasted content Class1 >; Combined information also can be < user interest i, broadcast content identification j> etc.
Step S205, searches the fixed reference feature matched with described combined information in described first group of fixed reference feature and described second group of fixed reference feature.
In one embodiment, after user profile and content information are combined into different combined informations, the fixed reference feature with various combination information match can be searched respectively.After the user property in fixed reference feature and broadcasted content attribute kit are containing the user profile in combined information and content information, can determine that this fixed reference feature is the fixed reference feature needed.
Concrete, can search successively multiple candidate broadcast content, namely search the fixed reference feature that the content information of a certain candidate broadcast content is corresponding, after finding, the fixed reference feature that next candidate broadcast content matches can be searched successively.
Step S206, the eigenwert of the fixed reference feature found described in acquisition.
In one embodiment, when after the fixed reference feature finding coupling, obtain the eigenwert of the fixed reference feature found.As shown in the data layout of fixed reference feature that describes in above-described embodiment, when after the fixed reference feature finding Corresponding matching, can obtain eigenwert in data layout, if fixed reference feature is COEC feature, then characteristic of correspondence value is COEC value.
Step S207, according to described eigenwert, sorts described fixed reference feature.
In one embodiment, this step also can comprise:
According to the ratio of the described number of clicks in described eigenwert and described expected click number of times, what calculate described fixed reference feature estimates clicking rate;
Estimate clicking rate order from big to small according to described, described fixed reference feature is sorted.
In one embodiment, by gdbt pattern, fixed reference feature pattern is sorted.Concrete, can Node configuration threshold value in gdbt pattern, wherein, threshold value can be arranged according to COEC value, inputs in gdbt pattern with reference to the characteristic key words in feature and eigenwert simultaneously, gdbt pattern judges on node whether the eigenwert of fixed reference feature reaches the threshold value of setting, if reach the threshold value of setting, then show to estimate clicking rate high, then before this fixed reference feature can being come, concrete, this feature is located at the left branch of gdbt.By judging, can sort to fixed reference feature according to estimating clicking rate.
Optionally, the eigenwert of fixed reference feature also can comprise conversion number of times (x) etc.If eigenwert for transforming number of times, then can sort to fixed reference feature according to estimating conversion ratio.In N number of broadcasted content as the content source of some broadcasted contents, N number of broadcasted content can be sorted according to estimating conversion ratio, thus the high one or more broadcasted contents of sequence can be recommended user.
Step S208, chooses at least one fixed reference feature in described sequence.
Step S209, recommends user by broadcasted content corresponding for broadcasted content attribute in described fixed reference feature.
In one embodiment, after fixed reference feature is sorted, can choose according to putting in order and estimate clicking rate or estimate the high one or more fixed reference features of conversion ratio to recommend user.Concrete, can detect the terminal of report of user information, and promote broadcasted content corresponding to one or more fixed reference feature to this terminal, be animation form as comprised broadcasted content type in content information, then with animation form to this terminal recommending broadcast contents.
In the embodiment of the present invention, when after the content information getting user profile and candidate broadcast content, first group of fixed reference feature of the exterior storage relevant to described user profile and second group of fixed reference feature of storage inside can be obtained respectively, and the fixed reference feature by getting, search the fixed reference feature matched with user profile and content information, after finding corresponding fixed reference feature, can screen candidate broadcast content according to fixed reference feature, to filter out the broadcasted content high with user's degree of correlation.By being pre-stored in external memory storage and internal storage with reference to feature, can realizing once calling and user-dependent fixed reference feature from external memory storage, reducing the inquiry times to external memory storage, thus can inquiry time delay be reduced.
See Fig. 3, it is the structural representation of an embodiment of the recommendation apparatus of a kind of broadcasted content in the present invention.This device realizes the implementation method in above-described embodiment by concrete module.Concrete, this device can comprise: acquisition module 301, search module 302, screening module 303.
Wherein, acquisition module 301, for when after the content information getting user profile and candidate broadcast content, obtains first group of fixed reference feature of the exterior storage relevant to described user profile and second group of fixed reference feature of storage inside.
In one embodiment, after the broadcasted content in server recommends platform to obtain the content information of user profile and candidate broadcast content, can be one or more broadcasted contents in user's recommended candidate broadcasted content according to user profile and content information.Wherein, server obtains user profile etc. by terminal to report.Optionally, user profile can comprise the information such as user ID, age of user, sex, interest, city, user place, country and/or the interested city of user, country; Content information can comprise broadcast content identification, the content source mark of broadcasted content, popularization plan mark, broadcasted content classification, broadcasted content represent the information such as form, broadcasted content source.
After getting and in advance carrying out the user profile of recommending broadcast contents and the content information of multiple candidate broadcast content to it, acquisition module 301 can obtain first group of fixed reference feature of the exterior storage relevant to user profile and second group of fixed reference feature of storage inside.Concrete, when server by broadcasted content recommend platform carry out broadcasted content recommend time, the important references feature need recommended as broadcasted content by the fixed reference feature of server internal or outside pre-stored.Optionally, as being user's recommended advertisements, then fixed reference feature can be COEC feature, wherein, as the fixed reference features such as COEC feature can comprise characteristic key words and eigenwert, characteristic key words can comprise user property and broadcasted content attribute, and user property is corresponding with user profile, and broadcasted content attribute is corresponding with content information.Namely by searching the fixed reference feature of outside or inside pre-stored, obtain the eigenwert of the fixed reference feature matched with current user information and content information, then the contrast by eigenwert obtains one or more broadcasted content high with user's degree of correlation.
Optionally, during the fixed reference feature of exterior storage stores by Redis.Wherein, the internal memory that Redis can combine multiple machine forms a memory system, and supports with < characteristic key words, and the mode of eigenwert > stores.Wherein, the fixed reference feature that low frequency is inquired about and memory space is large can be stored in the external memory storage that server connects.The fixed reference feature comprising user ID as user property is stored in external memory storage, when being a certain user's recommending broadcast contents, from external memory storage, only transfers the fixed reference feature that the user ID of this user is relevant; And the fixed reference feature that other non-user mark user property can not being comprised user ID is correlated with is stored in internal storage, supports high frequency inquiry.
Search module 302, in the described first group fixed reference feature relevant to described user profile got and described second group of fixed reference feature, search the fixed reference feature matched with the content information in described candidate broadcast content.
In one embodiment, according to user profile, the first group of fixed reference feature matched with user profile can be obtained from external memory storage, and second group of fixed reference feature can be obtained from internal storage.Optionally, the fixed reference feature obtained can be placed in the buffer, thus can further improve search efficiency.After getting the fixed reference feature relevant to active user, search module 302 and search the fixed reference feature matched with the content information of current candidate broadcasted content.Concrete, if fixed reference feature is COEC feature, then the data layout of feature can be < user property, broadcasted content attribute, number of clicks (c), expected click number of times (ec), COEC>, wherein, number of clicks (c), expected click number of times (ec), COEC is the eigenwert of COEC feature.The COEC feature that the content search that then can comprise according to the content comprised in user profile and some content informations matches.If user ID is user 1, broadcasted content classification 1 is comprised in content information, then can search fixed reference feature is < user 1, broadcasted content classification, the fixed reference feature of broadcasted content classification 1 whether is had in c, ec, COEC>, if exist, then show that this fixed reference feature and content information and user profile match.
Optionally, matched and searched successively can be carried out to candidate broadcast content, as searched the fixed reference feature matched according to the content information of a certain broadcasted content, after finding the fixed reference feature that this broadcasted content matches, can continue to search next broadcasted content according to the same manner.
Screening module 303, for after finding the fixed reference feature matched with the content information of described candidate broadcast content, according to the described fixed reference feature found, filters out at least one broadcasted content to recommend user from described candidate broadcast content.
In one embodiment, when searching after module 302 finds the fixed reference feature matched with the content information of candidate broadcast content and user profile, screening module 303 can according to the fixed reference feature found, candidate broadcast content is screened, filters out the one or more broadcasted contents high with active user's degree of correlation to recommend user.Optionally, by iteration decision tree, (English: GradientBoostingDecisionTree is called for short: gdbt) model is to alternating content information sifting.By the eigenwert of the fixed reference feature of the coupling of acquisition, and Corresponding matching broadcasted content can be obtained according to this eigenwert estimate clicking rate, thus according to the size estimating clicking rate, candidate broadcast content can be sorted, recommend user by estimating the clicking rate high one or more broadcasted contents that sort.Optionally, according to the concrete form in content information, broadcasted content can be promoted to the terminal providing user profile.
In the embodiment of the present invention, when after the content information getting user profile and candidate broadcast content, first group of fixed reference feature of the exterior storage relevant to described user profile and second group of fixed reference feature of storage inside can be obtained respectively, and the fixed reference feature by getting, search the fixed reference feature matched with user profile and content information, after finding corresponding fixed reference feature, can screen candidate broadcast content according to fixed reference feature, to filter out the broadcasted content high with user's degree of correlation.By being pre-stored in external memory storage and internal storage with reference to feature, can realizing once calling and user-dependent fixed reference feature from external memory storage, reducing the inquiry times to external memory storage, thus can inquiry time delay be reduced.
See Fig. 4, it is the structural representation of another embodiment of the recommendation apparatus of a kind of broadcasted content in the present invention.This device realizes the implementation method in above-described embodiment by concrete module.Concrete, this device can comprise: acquisition module 401, search module 402, screening module 403.
Acquisition module 401, for when after the content information getting user profile and candidate broadcast content, obtains first group of fixed reference feature of the exterior storage relevant to described user profile and second group of fixed reference feature of storage inside.
In one embodiment, acquisition module 401 can comprise transmitting element 4011, first acquiring unit 4012, second acquisition unit 4013.
Wherein, transmitting element 4011, send inquiry request for externally storer, wherein, described inquiry request comprises the user ID in described user profile.
In one embodiment, after the broadcasted content in server recommends platform to obtain the content information of user profile and candidate broadcast content, can be one or more broadcasted contents in user's recommended candidate broadcasted content according to user profile and content information.Wherein, server obtains user profile etc. by terminal to report.Optionally, user profile can comprise the information such as user ID, age of user, sex, interest, city, user place, country and/or the interested city of user, country; Content information can comprise broadcast content identification, the content source mark of broadcasted content, popularization plan mark, broadcasted content classification, broadcasted content represent the information such as form, broadcasted content source.As being user's recommended advertisements, then fixed reference feature can be COEC feature, wherein, as the fixed reference features such as COEC feature can comprise characteristic key words and eigenwert, characteristic key words can comprise user property and broadcasted content attribute, user property is corresponding with user profile, and broadcasted content attribute is corresponding with content information.
In one embodiment, during the fixed reference feature of exterior storage stores by Redis.Wherein, the internal memory that Redis can combine multiple machine forms a memory system, and supports with < characteristic key words, and the mode of eigenwert > stores.After server gets user profile, user ID can be sent to external memory storage by query statement by transmitting element 4011, namely in Redis storer.Wherein, the fixed reference feature that user ID is relevant is stored in external memory storage.Due to user ID and customer volume linear, therefore in external memory storage, store most of fixed reference feature in fixed reference feature storehouse, inquiry request is sent by externally storer, disposablely can call fixed reference feature corresponding to active user, thus achieve memory space greatly and the effect that reduces of the number of times externally inquired about.
First acquiring unit 4012, for obtaining first group of fixed reference feature of described external memory storage feedback, wherein, the user property in described first group of fixed reference feature comprises described user ID.
In one embodiment, after external memory storage receives the inquiry request of server, the user ID in inquiry request can be parsed, and the fixed reference feature of this user ID can be comprised in inquiring user attribute, and the fixed reference feature comprising this user ID can be fed back to server with the form of group.After the first acquiring unit 4012 gets first group of fixed reference feature of external memory storage feedback, namely get the fixed reference feature relevant to user ID.If one of them fixed reference feature in retrievable first group of fixed reference feature is < user ID 1, broadcast content identification k, broadcasted content type m, c, ec, COEC> etc.
Second acquisition unit 4013, for obtaining second group of fixed reference feature from internal storage, wherein, the user property in described second group of fixed reference feature comprises the user profile except described user ID.
In one embodiment, other user profile except user ID in user profile, also can be comprised, as user interest etc., then the fixed reference feature that the non-user mark that second acquisition unit 4013 can obtain pre-stored from internal storage is correlated with.If one of them fixed reference feature in retrievable second group of fixed reference feature is < user interest n, broadcast content identification i, content source mark j, c, ec, the COEC> etc. of broadcasted content.Wherein, internal storage can back-level server high frequency inquiry, as user interest can comprise P, broadcasted content type comprises Q, when needing to search the fixed reference feature all mated user interest and broadcasted content type, need inquiry P*Q time, enquiry frequency is higher, but can realize the low effect of inquiry time delay in internal storage.
Search module 402, in the described first group fixed reference feature relevant to described user profile got and described second group of fixed reference feature, search the fixed reference feature matched with the content information in described candidate broadcast content.
In one embodiment, search module 402 can comprise assembled unit 4021, search unit 4022.
Wherein, assembled unit 4021, for being undertaken combining to form combined information by the content information of described user profile and described candidate broadcast content.
In one embodiment, after getting the fixed reference feature relevant to user profile, whether assembled unit 4021, namely by being combined by the content information of user profile and candidate broadcast content, exists the fixed reference feature mated with combined information to search further in the first group of fixed reference feature got and second group of fixed reference feature.Illustrate, data layout due to fixed reference feature is < user property, broadcasted content attribute, eigenwert >, the data layout of combined information can be set to < user profile, content information >.Wherein, if according to the difference of user profile and content information, may be combined with into different combined informations.As combined information can be < user ID 1, broadcasted content Class1 >; Combined information also can be < user interest i, broadcast content identification j> etc.
Search unit 4022, for searching the fixed reference feature matched with described combined information in described first group of fixed reference feature and described second group of fixed reference feature.
In one embodiment, after user profile and content information are combined into different combined informations by assembled unit 4021, search unit 4022 and can search the fixed reference feature with various combination information match respectively.After the user property in fixed reference feature and broadcasted content attribute kit are containing the user profile in combined information and content information, can determine that this fixed reference feature is the fixed reference feature needed.
Concrete, searching unit 4022 can search successively to multiple candidate broadcast content, namely searches the fixed reference feature that the content information of a certain candidate broadcast content is corresponding, after finding, can search the fixed reference feature that next candidate broadcast content matches successively.
Screening module 403, for after finding the fixed reference feature matched with the content information of described candidate broadcast content, according to the described fixed reference feature found, filters out at least one broadcasted content to recommend user from described candidate broadcast content.
In one embodiment, screen module 403 can comprise the 3rd acquiring unit 4031, sequencing unit 4032, choose unit 4033, recommendation unit 4044.
Wherein, the 3rd acquiring unit 4031, for the eigenwert of fixed reference feature found described in obtaining.
In one embodiment, after searching unit 4022 and finding the fixed reference feature of coupling, the 3rd acquiring unit 4031 obtains the eigenwert of the fixed reference feature found.As shown in the data layout of fixed reference feature that describes in above-described embodiment, when after the fixed reference feature finding Corresponding matching, can obtain eigenwert in data layout, if fixed reference feature is COEC feature, then characteristic of correspondence value is COEC value.
Sequencing unit 4032, for according to described eigenwert, sorts described fixed reference feature.
In one embodiment, sequencing unit 4032 also can comprise following subelement:
Computation subunit, for the ratio according to the described number of clicks in described eigenwert and described expected click number of times, what calculate described fixed reference feature estimates clicking rate;
Sequence subelement, for estimating clicking rate order from big to small according to described, sorts to described fixed reference feature.
In one embodiment, the subelement that sorts sorts to fixed reference feature pattern by gdbt pattern.Concrete, can Node configuration threshold value in gdbt pattern, wherein, threshold value can be arranged according to COEC value, inputs in gdbt pattern with reference to the characteristic key words in feature and eigenwert simultaneously, gdbt pattern judges on node whether the eigenwert of fixed reference feature reaches the threshold value of setting, if reach the threshold value of setting, then show to estimate clicking rate high, then before this fixed reference feature can being come, concrete, this feature is located at the left branch of gdbt.By judging, can sort to fixed reference feature according to estimating clicking rate.
Optionally, the eigenwert of fixed reference feature also can comprise conversion number of times (x) etc.If eigenwert for transforming number of times, then can sort to fixed reference feature according to estimating conversion ratio.In N number of broadcasted content as the content source of some broadcasted contents, N number of broadcasted content can be sorted according to estimating conversion ratio, thus the high one or more broadcasted contents of sequence can be recommended user.
Choose unit 4033, for choosing at least one fixed reference feature in described sequence.
Recommendation unit 4034, for recommending user by broadcasted content corresponding for broadcasted content attribute in described fixed reference feature.
In one embodiment, after fixed reference feature is sorted, choose unit 4033 and can choose according to putting in order and estimate clicking rate or estimate the high one or more fixed reference features of conversion ratio, and recommend user by recommendation unit 4034.Concrete, can detect the terminal of report of user information, and promote broadcasted content corresponding to one or more fixed reference feature to this terminal, be animation form as comprised broadcasted content type in content information, then with animation form to this terminal recommending broadcast contents.
In the embodiment of the present invention, when after the content information getting user profile and candidate broadcast content, first group of fixed reference feature of the exterior storage relevant to described user profile and second group of fixed reference feature of storage inside can be obtained respectively, and the fixed reference feature by getting, search the fixed reference feature matched with user profile and content information, after finding corresponding fixed reference feature, can screen candidate broadcast content according to fixed reference feature, to filter out the broadcasted content high with user's degree of correlation.By being pre-stored in external memory storage and internal storage with reference to feature, can realizing once calling and user-dependent fixed reference feature from external memory storage, reducing the inquiry times to external memory storage, thus can inquiry time delay be reduced.
See Fig. 5, it is the structural representation of the commending system of a kind of broadcasted content in the present invention.Wherein, this broadcasted content commending system comprises and estimates server 1 and external memory storage 2.As shown in the figure, estimate server and can comprise transceiver 11, communication bus 12, internal storage 13 and processor 14.Wherein, external memory storage 2 can be redis storage system, can store the reference data relevant to user ID that data volume is large, estimate server 1 to communicate with external memory storage by transceiver, the fixed reference feature relevant to user ID of external memory storage 2 storage is called by transceiver 11, transceiver 11 can comprise radio receiving transmitting module and wired transceiver module, communication bus 12 is for realizing the connection communication between these assemblies, internal storage 13 can be high-speed RAM storer, also can be non-labile storer (non-volatilememory), as at least one magnetic disk memory, wherein, internal storage can store the relevant fixed reference feature of non-user mark, also batch processing code can be stored, processor is by calling program code, the corresponding method shown in embodiment of Fig. 1 ~ Fig. 2 can be realized.Concrete, processor also can comprise the module that device described in Fig. 3 ~ Fig. 4 comprises.
In the embodiment of the present invention, by distinguishing with user ID with reference to feature, be pre-stored in external memory storage and internal storage respectively, external memory storage can support that memory space is large, and internal storage can support that high frequency is inquired about, thus the inquiry time delay of broadcasted content commending system can be reduced.
Device embodiment described above is only schematic, the wherein said unit illustrated as separating component or can may not be and physically separates, parts as unit display can be or may not be physical location, namely can be positioned at a place, or also can be distributed in multiple network element.Some or all of module wherein can be selected according to the actual needs to realize the object of the present embodiment scheme.Those of ordinary skill in the art, when not paying performing creative labour, are namely appreciated that and implement.
Step in embodiment of the present invention method can be carried out order according to actual needs and be adjusted, merges and delete.
Unit in embodiment of the present invention terminal or equipment or subelement can carry out merging, divide and deleting according to actual needs.
Through the above description of the embodiments, those skilled in the art can be well understood to the mode that each embodiment can add required general hardware platform by software and realize, and can certainly pass through hardware.Based on such understanding, technique scheme can embody with the form of software product the part that prior art contributes in essence in other words, this computer software product can store in a computer-readable storage medium, as ROM/RAM, magnetic disc, CD etc., comprising some instructions in order to make a computer equipment (can be personal computer, server, or the network equipment etc.) perform the method described in some part of each embodiment or embodiment.
Above-described embodiment, does not form the restriction to this technical scheme protection domain.The amendment done within any spirit at above-mentioned embodiment and principle, equivalently to replace and improvement etc., within the protection domain that all should be included in this technical scheme.
Claims (10)
1. a recommend method for broadcasted content, is characterized in that, comprising:
When after the content information getting user profile and candidate broadcast content, obtain first group of fixed reference feature of the exterior storage relevant to described user profile and second group of fixed reference feature of storage inside;
In the described first group fixed reference feature relevant to described user profile got and described second group of fixed reference feature, search the fixed reference feature matched with the content information of described candidate broadcast content;
After finding the fixed reference feature matched with the content information of described candidate broadcast content, according to the described fixed reference feature found, from described candidate broadcast content, filter out at least one broadcasted content to recommend user.
2. method as claimed in claim 1, it is characterized in that, described fixed reference feature comprises characteristic key words and eigenwert; Wherein,
Described characteristic key words comprises user property and broadcasted content attribute, and wherein, described user property comprises user ID;
Described eigenwert at least comprises number of clicks and expected click number of times.
3. method as claimed in claim 2, it is characterized in that, first group of fixed reference feature of the exterior storage that described acquisition is relevant to described user profile and second group of fixed reference feature of storage inside, comprising:
Externally storer sends inquiry request, and wherein, described inquiry request comprises the user ID in described user profile;
Obtain first group of fixed reference feature of described external memory storage feedback, wherein, the user property in described first group of fixed reference feature comprises described user ID;
From internal storage, obtain second group of fixed reference feature, wherein, the user property in described second group of fixed reference feature comprises the user profile except described user ID.
4. method as claimed in claim 3, it is characterized in that, in described described first group of fixed reference feature relevant to described user profile getting and described second group of fixed reference feature, searching the fixed reference feature matched with the content information of described candidate broadcast content, comprising:
The content information of described user profile and described candidate broadcast content is carried out combining to form combined information;
The fixed reference feature matched with described combined information is searched in described first group of fixed reference feature and described second group of fixed reference feature.
5., as method as described in arbitrary in claim 2-4, it is characterized in that, the fixed reference feature found described in described basis, from described candidate broadcast content, filter out at least one broadcasted content to recommend user, comprising:
The eigenwert of the fixed reference feature found described in acquisition;
According to described eigenwert, described fixed reference feature is sorted;
Choose at least one fixed reference feature in described sequence;
Broadcasted content corresponding for broadcasted content attribute in described fixed reference feature is recommended user.
6. a recommendation apparatus for broadcasted content, is characterized in that, comprising:
Acquisition module, for when after the content information getting user profile and candidate broadcast content, obtains first group of fixed reference feature of the exterior storage relevant to described user profile and second group of fixed reference feature of storage inside;
Search module, in the described first group fixed reference feature relevant to described user profile got and described second group of fixed reference feature, search the fixed reference feature matched with the content information of described candidate broadcast content;
Screening module, for after finding the fixed reference feature matched with the content information of described candidate broadcast content, according to the described fixed reference feature found, filters out at least one broadcasted content to recommend user from described candidate broadcast content.
7. device as claimed in claim 6, it is characterized in that, described fixed reference feature comprises characteristic key words and eigenwert; Wherein,
Described characteristic key words comprises user property and broadcasted content attribute, and wherein, described user property comprises user ID;
Described eigenwert at least comprises number of clicks and expected click number of times.
8. device as claimed in claim 7, it is characterized in that, described acquisition module comprises:
Transmitting element, send inquiry request for externally storer, wherein, described inquiry request comprises the user ID in described user profile;
First acquiring unit, for obtaining first group of fixed reference feature of described external memory storage feedback, wherein, the user property in described first group of fixed reference feature comprises described user ID;
Second acquisition unit, for obtaining second group of fixed reference feature from internal storage, wherein, the user property in described second group of fixed reference feature comprises the user profile except described user ID.
9. device as claimed in claim 8, is characterized in that, described in search module and comprise:
Assembled unit, for being undertaken combining to form combined information by the content information of described user profile and described candidate broadcast content;
Search unit, for searching the fixed reference feature matched with described combined information in described first group of fixed reference feature and described second group of fixed reference feature.
10. device as described in as arbitrary in claim 7-9, it is characterized in that, described screening module comprises:
3rd acquiring unit, for the eigenwert of fixed reference feature found described in obtaining;
Sequencing unit, for according to described eigenwert, sorts described fixed reference feature;
Choose unit, for choosing at least one fixed reference feature in described sequence;
Recommendation unit, for recommending user by broadcasted content corresponding for broadcasted content attribute in described fixed reference feature.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510595693.8A CN105224617A (en) | 2015-09-17 | 2015-09-17 | Recommendation method, device and system for broadcast content |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510595693.8A CN105224617A (en) | 2015-09-17 | 2015-09-17 | Recommendation method, device and system for broadcast content |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105224617A true CN105224617A (en) | 2016-01-06 |
Family
ID=54993585
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510595693.8A Pending CN105224617A (en) | 2015-09-17 | 2015-09-17 | Recommendation method, device and system for broadcast content |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105224617A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111224731A (en) * | 2019-12-26 | 2020-06-02 | 支付宝(杭州)信息技术有限公司 | Content pushing method, device and equipment based on voice broadcast |
| CN112232857A (en) * | 2020-09-25 | 2021-01-15 | 上海淇毓信息科技有限公司 | Automatic advertisement sequencing method and device and electronic equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101588216A (en) * | 2008-05-21 | 2009-11-25 | 华为技术有限公司 | Method, device and system for realizing media content playing in an interrupting way by internet protocol (IP) television |
| CN104301758A (en) * | 2014-10-10 | 2015-01-21 | 安徽华米信息科技有限公司 | Method, device and system for pushing videos |
-
2015
- 2015-09-17 CN CN201510595693.8A patent/CN105224617A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101588216A (en) * | 2008-05-21 | 2009-11-25 | 华为技术有限公司 | Method, device and system for realizing media content playing in an interrupting way by internet protocol (IP) television |
| CN104301758A (en) * | 2014-10-10 | 2015-01-21 | 安徽华米信息科技有限公司 | Method, device and system for pushing videos |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111224731A (en) * | 2019-12-26 | 2020-06-02 | 支付宝(杭州)信息技术有限公司 | Content pushing method, device and equipment based on voice broadcast |
| CN111224731B (en) * | 2019-12-26 | 2021-06-08 | 支付宝(杭州)信息技术有限公司 | Content pushing method, device and equipment based on voice broadcast |
| CN112232857A (en) * | 2020-09-25 | 2021-01-15 | 上海淇毓信息科技有限公司 | Automatic advertisement sequencing method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10789311B2 (en) | Method and device for selecting data content to be pushed to terminal, and non-transitory computer storage medium | |
| CN107526807B (en) | Information recommendation method and device | |
| US9087108B2 (en) | Determination of category information using multiple stages | |
| CN105653675A (en) | Information recommendation method and apparatus, server and system | |
| CN102402605A (en) | Mixed distribution model for search engine indexing | |
| CN104504077A (en) | Web access data statistical method and the device | |
| CN103327049B (en) | Based on rich content delivery method and the system of browser address bar | |
| KR101874862B1 (en) | Intelligent Searching System for Billing data of Internet service and Method thereof | |
| CN103198067A (en) | Business searching method and system | |
| CN103365904A (en) | Advertising information searching method and system | |
| CN104462396A (en) | Method and device for handing character strings | |
| KR20180017085A (en) | Location information providing method and device | |
| CN104484406A (en) | Recommended data issuing analysis method and recommended data issuing analysis device | |
| CN105808642B (en) | Recommendation method and device | |
| CN111461754A (en) | Method and device for determining flow source of order | |
| CN114265927A (en) | Data query method and device, storage medium and electronic device | |
| CN101694656A (en) | Search request method, search method, device and system | |
| CN106649385B (en) | Data reordering method and device based on HBase database | |
| CN108416645B (en) | Recommendation method, device, storage medium and equipment for user | |
| CN105224617A (en) | Recommendation method, device and system for broadcast content | |
| CN113139113A (en) | Search request processing method and device | |
| CN113961693A (en) | Search result recommendation method and device, electronic device and readable storage medium | |
| US20210035167A1 (en) | System and method for recommending digital advertisements and publishers | |
| CN113327146A (en) | Information tracking method and device | |
| CN113505295A (en) | Enterprise customer acquisition push algorithm implementation method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160106 |
|
| RJ01 | Rejection of invention patent application after publication |