CN105868264A - Method and system for pushing update information according to focus word - Google Patents
Method and system for pushing update information according to focus word Download PDFInfo
- Publication number
- CN105868264A CN105868264A CN201511032246.8A CN201511032246A CN105868264A CN 105868264 A CN105868264 A CN 105868264A CN 201511032246 A CN201511032246 A CN 201511032246A CN 105868264 A CN105868264 A CN 105868264A
- Authority
- CN
- China
- Prior art keywords
- word
- user
- information
- content
- fresh information
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
Abstract
The present invention discloses a method and system for pushing update information according to a focus word. The method comprises: generating a recommended word list according to currently played content, and pushing the recommended word list to user; after receiving a recommended word selected by the user in the recommended word list, using the recommended word selected by the user as a focus word and recording the focus word; and when acquiring update information in a server message message queue, matching the update information with the focus word, and pushing the update information matched with the focus word to the user. Instant pushing of the update information is implemented, invalid searching of the user is reduced, and time is saved for the user. Moreover, the method and system disclosed by the present invention adopt an algorithm of calculating a matching value between the focus word and the update information, which makes the push result more accurate and provides update information with high attention for the user.
Description
Technical field
The present invention relates to multimedia technology field, particularly relate to a kind of method and system pushing more fresh information according to concern word.
Background technology
In recent years, network search engines technology has been achieved for the biggest progress, and up to the present a conjugally developed has three generations's search engine.First generation search engine relies primarily on the classified catalogue of manual sorting and scans for；Second filial generation search engine relies primarily on machine crawl and scans for, and it is built upon the Webpage search on the basis of Hypertext Link；Third generation search engine technique then scans for as principal character with computer with user under interaction is pointed out.Current search engine mainly performs Webpage search, and the resource of search is also concentrated mainly on the webpage on the Internet and the file being associated.Utilize search engine can obtain the information based on webpage of magnanimity, but while obtaining useful information, also bring a lot of junk information, and for not being published in the information on webpage in a large number, search engine is the most helpless.
Along with popularizing of the Internet, knowledge, content on the Internet the most increasingly present explosive growth.The Internet both brings the benefit having everything to users, brings the puzzlement having no way of selecting also to user.On the one hand, in the face of vast as the open sea information, the user of information takes a considerable amount of time and is also difficult to find that oneself required information with expense；On the other hand, the publisher of information also is difficult to be sent to information in time, on one's own initiative user interested.
Owing to user may be simultaneously interested in multiple things, and use search engine can only retrieve the relevant information of a point of interest, it is thus desirable to search is repeated, and when certain point of interest is needed to give more sustained attention by user, also must constantly repeat the network address searched or open in bookmark, process is more complicated；And use the mode subscription information such as RSS, and needing subscribing to the most respectively different point of interest, process is loaded down with trivial details, brings great inconvenience to user.
Summary of the invention
In view of this, it is an object of the invention to propose a kind of saving user time, it is possible to the more fresh information of information user paid close attention to is pushed to the method and system of user in time.
A kind of method pushing more fresh information according to concern word provided based on the above-mentioned purpose present invention, including:
Generate and recommend word list, be pushed to user；
Receive user to described recommendation word list is recommended the selection of word, and recommendation word user selected is as paying close attention to word record；
Obtain the more fresh information in server message queue, and mate with described concern word；
By the renewal information pushing that mates with described concern word to user.
Further, described by the renewal information pushing mated with described concern word to user during, also include calculating the matching value paying close attention to word and described more fresh information, it is judged that whether matching value reaches default threshold value, the most then be judged to coupling.
Further, described matching value is determined by the evaluation score of more fresh information, and the algorithm of evaluation score is as follows:
Calculate key word k matching value S (k, D) in up-to-date data item content,
S (k, D)=T (k, D) × Ik；
Wherein k pays close attention to word, and D represents that document, Ik represent inverse document word frequency；
T (k, D) pays close attention to word k word frequency in document D, is calculated by equation below:
T pays close attention to word k and derivative words thereof the occurrence number in the document D of up-to-date data item content, l represents the number recommending word in document D, dl is the average length of all documents in up-to-date data item content, s is system self-regulation integral coefficient, its span is (0,1), specifically jointly determined by document t, l and dl；
IkCalculated by equation below:
Df represents containing paying close attention to the frequency that word k occurs in all documents；
S (Q, D)=∑k ∈ QS (k, D)；
S (Q, D) is the sum of the matching value S (k, D) of all concern word k in set Q.
Further, described generation recommends the step of word list to include: according to the information of currently playing content, generates and recommends word list.
Further, described generation recommends the step of word list also to include: when changing broadcasting content, and the information playing content according to replacing generates new recommendation word list, and recommends the concern word in word list before interpolation in described new recommendation word list.
Further, when the more fresh information obtained in server message queue, the time of this operation is recorded；When the more fresh information again obtained in server message queue, at the time of described record, start to obtain more fresh information, the time of record before deletion, and record the new operating time.
Further, the more fresh information in described server message queue is to be generated according to the latest data item content obtained from data base by server, and the information being stored in message queue.
Further, described more fresh information includes the title of data item content, key word and broadcast address, if described data item content is video content, then described in more fresh information also include the character names in video poster picture, director names, protagonist name and video content；Described concern word includes title and the key word playing content, if described broadcasting content is video content, the most described concern word also includes the character names in video poster picture, director names, protagonist name and video content.
Further, during the renewal information pushing that will match with described concern word is to user, in described more fresh information is arranged in propelling movement list according to matching degree value order from high to low, it is pushed to user.
Based on above-mentioned purpose, the present invention also provides for a kind of system pushing more fresh information according to concern word, including:
Recommend word List Generating Module, be used for generating recommendation word list, be pushed to user；
Paying close attention to word setting module, for receiving user to recommending the selection of word in described recommendation word list, and recommendation word user selected is as paying close attention to word record；
Update data obtaining module, for obtaining the more fresh information in server message queue, and mate with described concern word；
Updating info push module, the renewal information pushing being used for mate with described concern word is to user.
Further, described renewal info push module by the renewal information pushing that mates with described concern word to user during, also include calculating the matching value paying close attention to word with described more fresh information, it is judged that whether matching value reaches default threshold value, the most then it is judged to coupling.
Further, described matching value is determined by the evaluation score of more fresh information, and the algorithm of evaluation score is as follows:
Calculate key word k matching value S (k, D) in up-to-date data item content,
S (k, D)=T (k, D) × Ik；
Wherein k pays close attention to word, and D represents that document, Ik represent inverse document word frequency；
T (k, D) pays close attention to word k word frequency in document D, is calculated by equation below:
T pays close attention to word k and derivative words thereof the occurrence number in the document D of up-to-date data item content, l represents the number recommending word in document D, dl is the average length of all documents in up-to-date data item content, s is system self-regulation integral coefficient, its span is (0,1), specifically jointly determined by document t, l and dl；
IkCalculated by equation below:
Df represents containing paying close attention to the frequency that word k occurs in all documents；
S (Q, D)=∑k ∈ QS (k, D)；
S (Q, D) is the sum of the matching value S (k, D) of all concern word k in set Q.
Further, described recommendation word List Generating Module generates and recommends the step of word list to include: according to the information of currently playing content, generates and recommends word list.
Further, described recommendation word List Generating Module generates recommends the step of word list also to include: when changing broadcasting content, the information playing content according to replacing generates new recommendation word list, and at the described new concern word recommending to recommend before interpolation in word list in word list.
Further, when the more fresh information obtained in server message queue, the time of described renewal data obtaining module this operation of record；When the more fresh information again obtained in server message queue, at the time of described record, start to obtain more fresh information, the time of record before deletion, and record the new operating time.
Further, the more fresh information in described server message queue is to be generated according to the latest data item content obtained from data base by server, and the information being stored in message queue.
Further, described more fresh information includes the title of data item content, key word and broadcast address, if described data item content is video content, then described in more fresh information also include the character names in video poster picture, director names, protagonist name and video content；Described concern word includes title and the key word playing content, if described broadcasting content is video content, the most described concern word also includes the character names in video poster picture, director names, protagonist name and video content.
Further, described more new information push by the renewal information pushing that matches with described concern word to user during, be pushed to user in described more fresh information is arranged in propelling movement list according to matching degree value order from high to low.
From the above it can be seen that a kind of method and system pushing more fresh information according to concern word that the present invention provides, it is possible to generate according to currently playing content and recommend word list, be pushed to user；After recommending the selection of word in reception user is to described recommendation word list, recommendation word user selected is as paying close attention to word record；When obtaining the more fresh information in server message queue, described more fresh information is mated with described concern word, and by the renewal information pushing that mates with described concern word to user.Achieve the instant propelling movement of more fresh information, decrease the invalid search of user, saved the time for user.What the present invention used simultaneously calculates the method paid close attention to word with update information matches value so that push result more accurate.
Accompanying drawing explanation
Fig. 1 is that the present invention is according to paying close attention to the flow chart that word pushes a kind of embodiment of renewal information approach；
Fig. 2 is that the present invention is according to a kind of structure chart implementing to illustrate paying close attention to word propelling movement renewal information system；
Fig. 3 is that the present invention is according to paying close attention to the particular flow sheet that word pushes a kind of embodiment of renewal information approach.
Detailed description of the invention
For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with specific embodiment, and referring to the drawings, the present invention is described in more detail.
As it is shown in figure 1, be the present invention flow chart according to a kind of embodiment paying close attention to word propelling movement renewal information approach, including:
Step 101: generate according to currently playing content and recommend word list, and the recommendation word list of generation is pushed to user, described recommendation word list includes playing the key word of content, title, if described broadcasting content is video content, the most described recommendation word list also includes video poster picture, director names and protagonist name, and the character names in video content；
Step 102: accept user's selecting in recommending row vocabulary and pay close attention to the operation of word, the concern word that record user selects, during selecting to pay close attention to word, user can manually input the concern word wanting to set, searching the concern word wanting to set in recommending word list, user can also manually input setting and pay close attention to word simultaneously；
Step 103: obtain the more fresh information in server message queue, is mated with the more fresh information of acquisition by the word of paying close attention to of record in step 102；
Step 104: by the renewal information pushing that mates with described concern word to user.
As one embodiment of the present of invention, the generation of described recommendation list can also recommend word list by generating according to currently playing content, and user in historical record is selected pay close attention to word join generation recommend word list forms new recommendation word list.
As an alternative embodiment of the invention, the operation deleting concern word according to user, delete recommending the concern word being set in word list, will user delete concern word be reduced to recommend word, this recommendation word is not paid close attention to by user, this recommendation selected ci poem can also be selected as paying close attention to word by user again, then be again recorded as this recommendation word paying close attention to word in recommending word list.
In some embodiments of the invention, the information pushing that updates mated with described concern word is also included to the process of user: calculate the matching value paying close attention to word with described more fresh information, it is judged that whether matching value reaches default threshold value, the most then be judged to coupling.Wherein the algorithm of matching value is as follows,
Described matching value is determined by the evaluation score of more fresh information, and the algorithm of evaluation score is as follows:
Calculate key word k matching value S (k, D) in up-to-date data item content,
S (k, D)=T (k, D) × Ik；
Wherein k pays close attention to word, and D represents that document, Ik represent inverse document word frequency；
T (k, D) pays close attention to word k word frequency in document D, is calculated by equation below:
T pays close attention to word k and derivative words thereof the occurrence number in the document D of up-to-date data item content, l represents the number recommending word in document D, dl is the average length of all documents in up-to-date data item content, s is system self-regulation integral coefficient, its span is (0,1), specifically jointly determined by document t, l and dl；
IkCalculated by equation below:
Df represents containing paying close attention to the frequency that word k occurs in all documents；
S (Q, D)=∑k ∈ QS (k, D)；
S (Q, D) is the sum of the matching value S (k, D) of all concern word k in set Q.
The matching value of more fresh information is obtained according to above-mentioned algorithm, the matching value of described more fresh information is made comparisons with presetting threshold value, if the matching value of described more fresh information exceedes threshold value set in advance, then this broadcast information is added to pushing in list, if the matching value of described more fresh information is less than threshold value set in advance, then ignore this more fresh information, continue the size of matching value and the threshold value set in advance contrasting remaining more fresh information.If the matching value entirety of more fresh information is less than normal, then threshold value set in advance is turned down, if the matching value entirety of more fresh information is bigger than normal, then threshold value set in advance is tuned up, both lacked in will not having made propelling movement list and push content, push content in propelling movement list will not be made again too much.
As one embodiment of the present of invention, will update information pushing to user during, according to the calculated matching value of above-mentioned algorithm, more fresh information is arranged in recommendation list according to matching value order from big to small.
As an alternative embodiment of the invention, during the more fresh information in obtaining server message queue, record the time of this operation；When the more fresh information again obtained in server message queue, at the time of described record, start to obtain more fresh information, the time of record before deletion, and record the new operating time.
As another embodiment of the present invention, during the more fresh information in obtaining server message queue, it is also possible at the end of the last item more fresh information obtained, the correspondence position in server message queue adds label.When the more fresh information again obtained in server message queue, at the label of described interpolation, start to obtain more fresh information, the label added before deletion, and the correspondence position in server message queue adds label at the end of the last item got more fresh information.
As in figure 2 it is shown, be that the present invention is according to paying close attention to the particular flow sheet that word pushes a kind of embodiment of renewal information approach.
Described as follows according to the flow process paying close attention to word propelling movement renewal information approach:
Server obtains data item content up-to-date in data base, generates more fresh information according to described up-to-date data item content, and described more fresh information is stored in message queue.
Step 201: when user uses mobile terminal playing video content or audio content, described mobile terminal (i.e. user side) generates according to the information of currently playing content recommends word list.The information of described broadcasting content includes the character names in video content or the key word of audio content, title, and video poster picture, director names, protagonist name and video content.Described recommendation word list also includes the derivative words that user side derives according to the information playing content, if video content is " the Mi month passes ", the information then playing content includes the Mi month, Sun Li etc., then derivative words can be discriminate, film that Zhen Chuan and grandson pari performed or TV play, such as " machinery is chivalrous " or " Hydrargyri Oxydum Rubrum family " etc..
Step 202: recommendation word list is recommended user by user side, pays close attention to word according to user to the selection operation note recommending word.When user selects to recommend word in the recommendation word list that user side pushes, user directly can click on correspondence in recommending word list and recommend the selection icon of word, lookup can also be manually entered in recommending word list and recommend word, Query Result selects recommend word, additionally, user can also be manually entered in recommending word list and create key word.User side operates according to the selection of user, the recommendation word that record is chosen, and the word of recommending user selected is recorded as the concern word of corresponding user.
Step 203: data content up-to-date in server obtains data base, generates more fresh information, and is stored in message queue.
Step 204: user side can obtain more fresh information from message queue, record obtains the time of operation.Described more fresh information includes the title of data item content, key word and broadcast address, if described data item content is video content, then described in more fresh information also include the character names in video poster picture, director names, protagonist name and video content.
By the more fresh information got and user, concern word set in advance in recommending word list contrasts step 205 user side, and calculates the matching value paying close attention to word with more fresh information.Will character string in more fresh information with pay close attention to the character string contrast of word, calculate the evaluation score of more fresh information, evaluation score algorithm is as follows:
Described matching value is determined by the evaluation score of more fresh information, and the algorithm of evaluation score is as follows:
Calculate key word k matching value S (k, D) in up-to-date data item content,
S (k, D)=T (k, D) × Ik；
Wherein k pays close attention to word, and D represents that document, Ik represent inverse document word frequency；
T (k, D) pays close attention to word k word frequency in document D, is calculated by equation below:
T pays close attention to word k and derivative words thereof the occurrence number in the document D of up-to-date data item content, l represents the number recommending word in document D, dl is the average length of all documents in up-to-date data item content, the mean number recommending word in the most all documents, s is system self-regulation integral coefficient, its span is (0,1), is specifically jointly determined by document t, l and dl；
IkCalculated by equation below:
Df represents containing paying close attention to the frequency that word k occurs in all documents；
S (Q, D)=∑k ∈ QS (k, D)；
S (Q, D) is the sum of the matching value S (k, D) of all concern word k in set Q.
After calculating the matching value paying close attention to word and more fresh information, compared paying close attention to the word threshold value with the matching value of more fresh information and matching value set in advance by user side, if the matching value paying close attention to word and more fresh information exceedes threshold value set in advance, then push this more fresh information, if the matching value paying close attention to word and more fresh information is less than threshold value set in advance, the most do not push this more fresh information.In practice, if it is overall with the matching value of more fresh information bigger than normal or overall less than normal to pay close attention to word, then user terminal system adjusts threshold value automatically.When described matching value entirety is bigger than normal, then increase threshold value；When described matching value entirety is less than normal, then reduce threshold value.When paying close attention to the matching value skewness of word and more fresh information, then the coefficient s during user terminal system adjusts evaluation score algorithm automatically, the matching value of the change of directivity more fresh information.
Step 206: when user side filters out after matching value exceedes and preset the more fresh information of threshold value, more fresh information is added to recommendation list, compare the matching value paying close attention to word with more fresh information.
Step 207: user side will be in more fresh information be arranged in recommendation list according to matching value order from high to low.
Step 208: when receiving the play operation of user, when playing update content, the more fresh information during i.e. user clicks on recommendation list, obtain the broadcast address of more fresh information, and log in this broadcast address viewing update content.User side generates according to the information of currently playing content recommends word, and adds recommend word and the word of paying close attention to that user sets before that generate to new recommendation word list.
Step 209: when receiving user and the recommendation word in this recommendation word list carrying out selecting operation, record this selection operation, it is recorded as paying close attention to word and adding to the center of interest by corresponding recommendation word, when receiving user to the deletion action of the concern word in the center of interest, delete concern word corresponding in the center of interest.The concern word deleted from the center of interest can heavily be newly added in the center of interest.
Step 210: when user side obtains more fresh information again from the message queue of server, more fresh information can be obtained at the time of record, owing to the data content in server is to constantly update, but user side from the message team of server in obtain more fresh information be discontinuity.After getting more fresh information, can delete the temporal information recorded before, this time of new record of laying equal stress on obtains and updates the time that letter operation is corresponding.Afterwards, the concern word contrast in the center of interest again more fresh information got and user reset, calculate new more fresh information and the new matching value paying close attention to word.So circulation, allows users to obtain in time and pay close attention to the more fresh information that word mates, and then obtains the broadcast address of update content, and user is obtained by the broadcast address that login obtains and plays content, decreases the number of times of the invalid search of user, has saved the time for user.
Additionally provide a kind of system pushing more fresh information according to concern word in another aspect of this invention.
As shown in Figure 3, push a kind of structure chart implementing to illustrate of renewal information system according to concern word for the present invention, the system shown in figure includes recommending word List Generating Module 301, concern word setting module 302, updating data obtaining module 303 and renewal info push module 304.
Wherein, described recommendation word List Generating Module 301 is used for generating recommendation word list, and the recommendation word list of generation is pushed to user.During user side plays video content, word List Generating Module 301 is recommended to obtain the character names in the title of currently playing content, key word, video poster picture, director names, protagonist name and video content, and the relevant information such as Presence of the Moment, sheet caudal flexure, described information is derived, generate derivation information, and derivative as recommending word using the relevant information of video content and this relevant information, generate recommendation word list.
Described concern word setting module 302 is for receiving user to recommending the selection of word in described recommendation word list, and recommendation word user selected is as paying close attention to word record.When receiving user to the selection operation recommending word List Generating Module 301 generation to recommend to recommend in word list word, described concern word setting module 302 operates according to the selection of user, the recommendation word that record is corresponding is concern word, and concern word is added to the center of interest, described the center of interest is virtual state, and the most described the center of interest simply pays close attention to the set of word, additionally, described the center of interest can also be an independent list, and described list includes all of concern word.When receive user to the center of interest in pay close attention to the deletion action of word time, described concern word setting module 302 changes the selected state recommending to recommend word in word list, and is deleted by corresponding concern word in the center of interest.If described the center of interest is virtual state, the most described concern word setting module 302 only changes recommends the corresponding selected state recommending word in word list, will change free state into by the recommendation word selected of user.If described the center of interest is independent list, then in this independent list, concern word is removed.
Described renewal data obtaining module 303 is for obtaining the more fresh information in server message queue.When server count according to storehouse in have up-to-date data item content time, server generates more fresh information according to described up-to-date data item content, and described more fresh information is stored in message queue.Described renewal data obtaining module 303 obtains more fresh information from the message queue of server.Described more fresh information includes the title of data item content, key word and broadcast address, if described data item content is video content, then described in more fresh information also include the character names in video poster picture, director names, protagonist name and video content.During described renewal data obtaining module 303 obtains more fresh information, described renewal data obtaining module 303 records the time of this operation；When the more fresh information again obtained in server message queue, at the time of described record, start to obtain the time recorded before more fresh information, and deletion, again record this time and obtain the operating time of more fresh information.In addition, described renewal data obtaining module 303 is during obtaining more fresh information, can also add the label only oneself being capable of identify that in the message queue of server, i.e. at the end of the last item more fresh information obtained, the correspondence position in server message queue adds label.When the more fresh information again obtained in server message queue, at the label of described interpolation, start to obtain more fresh information, the label added before deletion, and label is added in the corresponding position in server message queue at the end of the last item got more fresh information.
The renewal information pushing that described renewal info push module 304 is used for mate with described concern word is to user.Described renewal info push module 304 is during the renewal information pushing got by described renewal data obtaining module 303 is to user, it is additionally operable to calculate the matching value of more fresh information and the user's concern set in advance word obtained, judges whether this renewal information pushing to user according to the size of matching value.The evaluation score of described more fresh information is determined by the size of described matching value by described renewal data obtaining module 303, and algorithm is as follows:
Calculate key word k matching value S (k, D) in up-to-date data item content,
S (k, D)=T (k, D) × Ik；
Wherein k pays close attention to word, and D represents that document, Ik represent inverse document word frequency；
T (k, D) pays close attention to word k word frequency in document D, is calculated by equation below:
T pays close attention to word k and derivative words thereof the occurrence number in the document D of up-to-date data item content, l represents the number recommending word in document D, dl is the average length of all documents in up-to-date data item content, s is system self-regulation integral coefficient, its span is (0,1), specifically jointly determined by document t, l and dl；
IkCalculated by equation below:
Df represents containing paying close attention to the frequency that word k occurs in all documents；
S (Q, D)=∑k ∈ QS (k, D)；
S (Q, D) is the sum of the matching value S (k, D) of all concern word k in set Q.
In addition, described renewal info push module 304 is during by the renewal information pushing that the match is successful to user, also include the sequence to the more fresh information that the match is successful, size according to calculated matching value, the more fresh information that the match is successful is arranged in propelling movement list according to updating the descending order of information matches value, and propelling movement list is pushed to user.The process of this propelling movement is: when user uses mobile terminal playing video content, the lower right corner of display screen or other do not affect user video and experience and compare again the most eye-catching region Pop-up message prompting frame, when receiving user to the clicking operation of this message notifying frame or drag operation, eject recommendation list.
Those of ordinary skill in the field are it is understood that the discussion of any of the above embodiment is exemplary only, it is not intended that hint the scope of the present disclosure (including claim) is limited to these examples；Under the thinking of the present invention, can also be combined between the technical characteristic in above example or different embodiment, and there is other change of many of the different aspect of the present invention as above, for they not offers in details simple and clear.Therefore, all within the spirit and principles in the present invention, any omission of being made, amendment, equivalent, improvement etc., should be included within the scope of the present invention.
Claims (18)
1. the method pushing more fresh information according to concern word, it is characterised in that including:
Generate and recommend word list, be pushed to user；
Receive user to described recommendation word list is recommended the selection of word, and recommendation word user selected is made
For paying close attention to word record；
Obtain the more fresh information in server message queue, and mate with described concern word；
By the renewal information pushing that mates with described concern word to user.
Method the most according to claim 1, it is characterised in that described will be with described concern word
During the renewal information pushing joined is to user, also include calculating pay close attention to word and described more fresh information
Join value, it is judged that whether matching value reaches default threshold value, the most then be judged to coupling.
Method the most according to claim 2, it is characterised in that described matching value is by more fresh information
Evaluation score determine, the algorithm of evaluation score is as follows:
Calculate key word k matching value S (k, D) in up-to-date data item content
S (k, D)=T (k, D) × Ik；
Wherein k pays close attention to word, and D represents that document, Ik represent inverse document word frequency；
T (k, D) pays close attention to word k word frequency in document D, is calculated by equation below:
T pays close attention to word k and derivative words thereof the occurrence number in the document D of up-to-date data item content,
L represents the number recommending word in document D, and dl is the average long of all documents in up-to-date data item content
Degree, s is system self-regulation integral coefficient, and its span is (0,1), specifically common by document t, l and dl
Determine；
IkCalculated by equation below:
Df represents containing paying close attention to the frequency that word k occurs in all documents；
S (Q, D)=∑k∈QS (k, D)；
S (Q, D) is the sum of the matching value S (k, D) of all concern word k in set Q.
Method the most according to claim 1, it is characterised in that the step of word list is recommended in described generation
Suddenly include: according to the information of currently playing content, generate and recommend word list.
Method the most according to claim 4, it is characterised in that the step of word list is recommended in described generation
Suddenly also include: when changing broadcasting content, generate new recommendation word row according to changing the information playing content
Table, and at the described new concern word recommending to recommend before interpolation in word list in word list.
Method the most according to claim 1, it is characterised in that when obtaining in server message queue
More fresh information time, record the time of this operation；When the renewal again obtained in server message queue
During information, at the time of described record, start to obtain more fresh information, the time of record before deletion, and
Record the new operating time.
Method the most according to claim 1, it is characterised in that in described server message queue
More fresh information is to be generated according to the latest data item content obtained from data base by server, and is stored in
Information in message queue.
Method the most according to claim 1, it is characterised in that described more fresh information includes data item
The title of content, key word and broadcast address, if described data item content is video content, then described in more
Fresh information also includes the character names in video poster picture, director names, protagonist name and video content；
Described concern word includes title and the key word playing content, if described broadcasting content is video content, then
Described concern word also includes the role in video poster picture, director names, protagonist name and video content
Name.
Method the most according to claim 1, it is characterised in that will match with described concern word
Renewal information pushing to user during, by described more fresh information according to matching degree value from high to low
Order be arranged in propelling movement list in be pushed to user.
10. the system pushing more fresh information according to concern word, it is characterised in that including:
Recommend word List Generating Module, be used for generating recommendation word list, be pushed to user；
Pay close attention to word setting module, for receiving user to described recommendation word list is recommended the selection of word, and
Recommendation word user selected is as paying close attention to word record；
Update data obtaining module, for obtaining the more fresh information in server message queue, and with described
Concern word mates；
Updating info push module, the renewal information pushing being used for mate with described concern word is to user.
11. systems according to claim 10, it is characterised in that at described renewal information pushing mould
Block by the renewal information pushing that mates with described concern word to user during, also include calculating paying close attention to word
Matching value with described more fresh information, it is judged that whether matching value reaches default threshold value, the most then judge
For coupling.
12. systems according to claim 11, it is characterised in that described matching value is by updating letter
The evaluation score of breath determines, the algorithm of evaluation score is as follows:
Calculate key word k matching value S (k, D) in up-to-date data item content
S (k, D)=T (k, D) × Ik；
Wherein k pays close attention to word, and D represents that document, Ik represent inverse document word frequency；
T (k, D) pays close attention to word k word frequency in document D, is calculated by equation below:
T pays close attention to word k and derivative words thereof the occurrence number in the document D of up-to-date data item content,
L represents the number recommending word in document D, and dl is the average long of all documents in up-to-date data item content
Degree, s is system self-regulation integral coefficient, and its span is (0,1), specifically common by document t, l and dl
Determine；
IkCalculated by equation below:
Df represents containing paying close attention to the frequency that word k occurs in all documents；
S (Q, D)=∑k∈QS (k, D)；
S (Q, D) is the sum of the matching value S (k, D) of all concern word k in set Q.
13. systems according to claim 10, it is characterised in that described recommendation word list generates mould
Block generates recommends the step of word list to include: according to the information of currently playing content, generates and recommends word list.
14. systems according to claim 10, it is characterised in that described recommendation word list generates mould
Block generates recommends the step of word list also to include: when changing broadcasting content, play content according to changing
Information generates new recommendation word list, and recommends word list in described new recommendation word list before interpolation
In concern word.
15. systems according to claim 10, it is characterised in that when obtaining server message queue
In more fresh information time, the time of described renewal data obtaining module this operation of record；When again obtaining
During more fresh information in server message queue, at the time of described record, start to obtain more fresh information,
The time of record before deletion, and record the new operating time.
16. systems according to claim 10, it is characterised in that in described server message queue
More fresh information be to be generated according to the latest data item content obtained from data base by server, and deposit
Enter the information in message queue.
17. systems according to claim 10, it is characterised in that described more fresh information includes data
The item title of content, key word and broadcast address, if described data item content is video content, then described
More fresh information also includes the role's surname in video poster picture, director names, protagonist name and video content
Name；Described concern word includes title and the key word playing content, if described broadcasting content is video content,
The most described concern word also includes the angle in video poster picture, director names, protagonist name and video content
Color name.
18. systems according to claim 10, it is characterised in that pushing at described more new information will
During the renewal information pushing matched with described concern word is to user, by described more fresh information according to
Matching degree value order from high to low is pushed to user in being arranged in propelling movement list.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511032246.8A CN105868264A (en) | 2015-12-31 | 2015-12-31 | Method and system for pushing update information according to focus word |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511032246.8A CN105868264A (en) | 2015-12-31 | 2015-12-31 | Method and system for pushing update information according to focus word |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105868264A true CN105868264A (en) | 2016-08-17 |
Family
ID=56624564
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511032246.8A Pending CN105868264A (en) | 2015-12-31 | 2015-12-31 | Method and system for pushing update information according to focus word |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105868264A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106484773A (en) * | 2016-09-12 | 2017-03-08 | 传线网络科技（上海）有限公司 | Determine the method and device of the weight of the keyword of multimedia resource |
| CN108491438A (en) * | 2018-02-12 | 2018-09-04 | 陆夏根 | A kind of technology policy retrieval analysis method |
| CN108683745A (en) * | 2018-05-24 | 2018-10-19 | 努比亚技术有限公司 | A kind of information updating and method for pushing |
| CN108737256A (en) * | 2018-06-13 | 2018-11-02 | 佛山市澄澜点寸科技有限公司 | It is a kind of to be used to receive the processing method and processing unit for sharing message |
| CN109522480A (en) * | 2018-11-12 | 2019-03-26 | 北京羽扇智信息科技有限公司 | A kind of information recommendation method, device, electronic equipment and storage medium |
| CN110177143A (en) * | 2019-05-27 | 2019-08-27 | 北京字节跳动网络技术有限公司 | Notification method, device, server and the readable medium that information updates |
| WO2019184111A1 (en) * | 2018-03-30 | 2019-10-03 | 平安科技（深圳）有限公司 | Push message processing method and apparatus, readable storage medium, and terminal device |
| CN111163348A (en) * | 2020-01-08 | 2020-05-15 | 百度在线网络技术（北京）有限公司 | Searching method and device based on video playing |
| CN112765514A (en) * | 2019-11-06 | 2021-05-07 | 腾讯科技（深圳）有限公司 | Method, device and storage medium for monitoring network public sentiment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101957857A (en) * | 2010-09-30 | 2011-01-26 | 华为终端有限公司 | Automatic information push method and server |
| CN102622445A (en) * | 2012-03-15 | 2012-08-01 | 华南理工大学 | User interest perception based webpage push system and webpage push method |
| CN102695086A (en) * | 2012-05-30 | 2012-09-26 | 杭州遥指科技有限公司 | Content pushing methods and device for interactive network protocol television |
| US20130151548A1 (en) * | 2011-12-07 | 2013-06-13 | Verizon Patent And Licensing Inc. | Media content searching |
| US20140165186A1 (en) * | 2012-12-12 | 2014-06-12 | Gowda Timma Ramu | Geometrical pattern based application passcode |
-
2015
- 2015-12-31 CN CN201511032246.8A patent/CN105868264A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101957857A (en) * | 2010-09-30 | 2011-01-26 | 华为终端有限公司 | Automatic information push method and server |
| US20130151548A1 (en) * | 2011-12-07 | 2013-06-13 | Verizon Patent And Licensing Inc. | Media content searching |
| CN102622445A (en) * | 2012-03-15 | 2012-08-01 | 华南理工大学 | User interest perception based webpage push system and webpage push method |
| CN102695086A (en) * | 2012-05-30 | 2012-09-26 | 杭州遥指科技有限公司 | Content pushing methods and device for interactive network protocol television |
| US20140165186A1 (en) * | 2012-12-12 | 2014-06-12 | Gowda Timma Ramu | Geometrical pattern based application passcode |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106484773A (en) * | 2016-09-12 | 2017-03-08 | 传线网络科技（上海）有限公司 | Determine the method and device of the weight of the keyword of multimedia resource |
| CN106484773B (en) * | 2016-09-12 | 2020-02-14 | 传线网络科技（上海）有限公司 | Method and device for determining weight of keyword of multimedia resource |
| CN108491438A (en) * | 2018-02-12 | 2018-09-04 | 陆夏根 | A kind of technology policy retrieval analysis method |
| WO2019184111A1 (en) * | 2018-03-30 | 2019-10-03 | 平安科技（深圳）有限公司 | Push message processing method and apparatus, readable storage medium, and terminal device |
| CN108683745A (en) * | 2018-05-24 | 2018-10-19 | 努比亚技术有限公司 | A kind of information updating and method for pushing |
| CN108737256A (en) * | 2018-06-13 | 2018-11-02 | 佛山市澄澜点寸科技有限公司 | It is a kind of to be used to receive the processing method and processing unit for sharing message |
| CN109522480A (en) * | 2018-11-12 | 2019-03-26 | 北京羽扇智信息科技有限公司 | A kind of information recommendation method, device, electronic equipment and storage medium |
| CN110177143A (en) * | 2019-05-27 | 2019-08-27 | 北京字节跳动网络技术有限公司 | Notification method, device, server and the readable medium that information updates |
| CN112765514A (en) * | 2019-11-06 | 2021-05-07 | 腾讯科技（深圳）有限公司 | Method, device and storage medium for monitoring network public sentiment |
| CN111163348A (en) * | 2020-01-08 | 2020-05-15 | 百度在线网络技术（北京）有限公司 | Searching method and device based on video playing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105868264A (en) | Method and system for pushing update information according to focus word | |
| US11606323B2 (en) | Prioritizing messages within a message network | |
| US10262045B2 (en) | Application representation for application editions | |
| US9317468B2 (en) | Personal content streams based on user-topic profiles | |
| US8181197B2 (en) | System and method for voting on popular video intervals | |
| US8903863B2 (en) | User interface with available multimedia content from multiple multimedia websites | |
| US9009589B2 (en) | Conversion of portable program modules for constrained displays | |
| US20170177572A1 (en) | Adaptive music and video recommendations | |
| CN105653572A (en) | Resource processing method and apparatus | |
| US20080294625A1 (en) | Item recommendation system | |
| US20130067348A1 (en) | Recommendation engine | |
| CN104813673A (en) | Sharing content-synchronized ratings | |
| WO2012173670A1 (en) | Systems and methods for providing media recommendations | |
| CN105245924B (en) | The video player of the push of video information and display methods and application this method | |
| JP2011135183A (en) | Information processing apparatus and method | |
| US8174579B2 (en) | Related scene addition apparatus and related scene addition method | |
| CN104798346A (en) | Enriching broadcast media related electronic messaging | |
| CN110430476A (en) | Direct broadcasting room searching method, system, computer equipment and storage medium | |
| KR101404596B1 (en) | System and method for providing video service based on image data | |
| KR20130083829A (en) | Automatic image discovery and recommendation for displayed television content | |
| US20130268261A1 (en) | Semantic enrichment by exploiting top-k processing | |
| US9542395B2 (en) | Systems and methods for determining alternative names | |
| CN104038790A (en) | Video playing list updating method and system | |
| US20170272793A1 (en) | Media content recommendation method and device | |
| JP5335500B2 (en) | Content search apparatus and computer program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160817 |