CN108073606A - A kind of news recommends method and apparatus, a kind of device recommended for news - Google Patents
A kind of news recommends method and apparatus, a kind of device recommended for news Download PDFInfo
- Publication number
- CN108073606A CN108073606A CN201610995502.1A CN201610995502A CN108073606A CN 108073606 A CN108073606 A CN 108073606A CN 201610995502 A CN201610995502 A CN 201610995502A CN 108073606 A CN108073606 A CN 108073606A
- Authority
- CN
- China
- Prior art keywords
- news
- primary keyword
- input content
- current input
- keyword
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
- G06F16/24578—Query processing with adaptation to user needs using ranking
-
- 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
Abstract
An embodiment of the present invention provides a kind of news to recommend method and apparatus, a kind of device recommended for news, and method therein specifically includes:Obtain the current input content of active user;When the current input content meets preset news correlated condition, the corresponding targeted news of news primary keyword are obtained, alternatively, obtaining news primary keyword and the corresponding targeted news of associative key;Wherein, the news primary keyword matches with the current input content, and the news primary keyword obtains for the history input behavior data according to multiple users, and the associative key is corresponding to the news primary keyword;Recommend the targeted news to the active user.The embodiment of the present invention can improve the accuracy of news recommendation, and can agree with the individual information needs of active user.
Description
Technical field
The present invention relates to technical field of information processing, more particularly to a kind of news method and apparatus, one kind are recommended to be used for
The device that news is recommended.
Background technology
With the fast development of internet, web database technology constantly increases, and offers convenience obtaining information to the network user
While also result in problem of information overload, how fast and effeciently to be searched and located in the data of magnanimity to the information needed into
For the developing outstanding problem of current internet.
To solve the above-mentioned problems, existing news website can select current hot topic on homepage or channel homepage
News, and be placed on comparison and wake up destination locations, user to be helped to find interested content, so as to effectively help user fast
Speed, the resource for correctly finding needs.
Inventor has found that the hot news that existing news website is recommended to different user is identical, however, interest is liked
It is good that individual reason is waited to cause different user there are different information requirements, in this way, the hot news that existing news website is recommended
The individual information needs for meeting user are will be unable to, cause the accuracy that news is recommended relatively low.For example, the user of the overwhelming majority is
In the news of concern " media event A ", but it is all interested in the news of " media event A " not represent all users, also
The user for having fraction is interested in other news, for example, it is interested in the news of " Olympic Games ", for another example, to the new of certain star
It hears interested etc., if recommending identical hot news to all user, will be unable to realize accurate recommendation, and can not meet
The individual information needs of user.
The content of the invention
In view of the above problems, the embodiment of the present invention proposes one kind and overcomes the above problem or solve at least partly above-mentioned
The news of problem recommends method, news recommendation apparatus and the device recommended for news, the embodiment of the present invention that can improve news
The accuracy of recommendation, and the individual information needs of active user can be agreed with.
To solve the above-mentioned problems, the invention discloses a kind of news to recommend method, including:
Obtain the current input content of active user;
When the current input content meets preset news correlated condition, the corresponding target of news primary keyword is obtained
News, alternatively, obtaining news primary keyword and the corresponding targeted news of associative key;Wherein, the news primary keyword with
The current input content matches, and the news primary keyword obtains for the history input behavior data according to multiple users,
The associative key is corresponding to the news primary keyword;
Recommend the targeted news to the active user.
Optionally, the current input content meets preset news correlated condition, including:
The current input content matches with news primary keyword;Or
The current input content matches with news primary keyword, and the current input content meets preset news
Statement pattern;Or
The current input content matches with news primary keyword, and the current input content and the main pass of the news
The corresponding associative key of keyword matches.
Optionally, news primary keyword is excavated as follows:
The vocabulary attribute of each vocabulary in the history input behavior data of multiple users is analyzed;
The vocabulary that vocabulary attribute meets preset attribute condition is chosen from the history input behavior data, as news master
Keyword.
Optionally, the vocabulary attribute includes:At least one of word frequency and number of users.
Optionally, the preset attribute condition includes:
Increment rate of the word frequency in the first preset time period is more than first threshold;And/or
Increment rate of the number of users in the second preset time period is more than second threshold.
Optionally, the method further includes:
Noise filtering is carried out to the news primary keyword by pseudo-linear filter mode.
Optionally, described the step of noise filtering is carried out to the news primary keyword by pseudo-linear filter mode, bag
It includes:
It searches for obtain corresponding first news search result according to the news primary keyword;
The corresponding associative key of the news primary keyword is excavated from first news search result;
It searches for obtain corresponding second news search result according to the associative key;
According to appearance information of the news primary keyword in second news search result, the news master is judged
Whether keyword is noise, if so, filtering out the news primary keyword.
Optionally, the corresponding associative key of the news primary keyword is excavated as follows:
The corresponding associative key of the news primary keyword is excavated from the history input behavior data of multiple users;
And/or
It searches for obtain corresponding news search result according to the news primary keyword, be dug from the news search result
Dig the corresponding associative key of the news primary keyword.
Optionally, described the step of obtaining the news primary keyword corresponding targeted news, including:
The corresponding targeted news of the news primary keyword are searched in news database;Or
Recommend to obtain the corresponding target of the news primary keyword in news list from the news primary keyword is corresponding
News.
On the other hand, the invention discloses a kind of news recommendation apparatus, including:
Input content acquisition module, for obtaining the current input content of active user;
News acquisition module, for when the current input content meets preset news correlated condition, obtaining news
The corresponding targeted news of primary keyword, alternatively, obtaining news primary keyword and the corresponding targeted news of associative key;Wherein,
The news primary keyword matches with the current input content, and the news primary keyword is the history according to multiple users
Input behavior data obtain, and the associative key is corresponding to the news primary keyword;And
News recommending module, for recommending the targeted news to the active user.
Optionally, the current input content meets preset news correlated condition, including:
The current input content matches with news primary keyword;Or
The current input content matches with news primary keyword, and the current input content meets preset news
Statement pattern;Or
The current input content matches with news primary keyword, and the current input content and the main pass of the news
The corresponding associative key of keyword matches.
Optionally, described device further includes:For excavating the first of news primary keyword the excavation module;
Described first excavates module, including:
Submodule is analyzed, the vocabulary attribute for each vocabulary in the history input behavior data to multiple users divides
Analysis;And
Submodule is chosen, meets preset attribute condition for choosing vocabulary attribute from the history input behavior data
Vocabulary, as news primary keyword.
Optionally, the vocabulary attribute includes:At least one of word frequency and number of users.
Optionally, the preset attribute condition includes:
Increment rate of the word frequency in the first preset time period is more than first threshold;And/or
Increment rate of the number of users in the second preset time period is more than second threshold.
Optionally, described device further includes:
Noise filtering module, for carrying out noise filtering to the news primary keyword by pseudo-linear filter mode.
Optionally, the noise filtering module includes:
First news search submodule, for searching for obtain corresponding first news search according to the news primary keyword
As a result;
Associative key excavates submodule, for excavating the news primary keyword from first news search result
Corresponding associative key;
Second news search submodule, for searching for obtain corresponding second news search knot according to the associative key
Fruit;
Judging submodule, for the believing in second news search result according to the news primary keyword
Breath, judges whether the news primary keyword is noise, if so, filtering out the news primary keyword.
Optionally, described device further includes:For excavating the second of the corresponding associative key of the news primary keyword
Excavate module;
Described second excavates module, including:
First excavates submodule, for excavating the news primary keyword from the history input behavior data of multiple users
Corresponding associative key;And/or
Second excavates submodule, for searching for obtain corresponding news search result according to the news primary keyword, from
The corresponding associative key of the news primary keyword is excavated in the news search result.
Optionally, the news acquisition module includes:
3rd news search submodule is new for searching for the corresponding target of the news primary keyword in news database
It hears;Or
List acquisition submodule, for recommending to obtain the news in news list from the news primary keyword is corresponding
The corresponding targeted news of primary keyword.
In another aspect, the invention discloses a kind of device recommended for news, include memory and one or
More than one program, either more than one program storage in memory and is configured to by one or one for one of them
A Yi Shang processor performs the one or more programs and includes the instruction operated below:
Obtain the current input content of active user;
When the current input content meets preset news correlated condition, the corresponding target of news primary keyword is obtained
News, alternatively, obtaining news primary keyword and the corresponding targeted news of associative key;Wherein, the news primary keyword with
The current input content matches, and the news primary keyword obtains for the history input behavior data according to multiple users,
The associative key is corresponding to the news primary keyword;
Recommend the targeted news to the active user.
The embodiment of the present invention includes advantages below:
The embodiment of the present invention is relevant with the presence or absence of the current input content to user by preset news correlated condition
News demand is identified exactly, can just be carried out when only the current input content meets preset news correlated condition new
It hears and recommends, also, the news primary keyword that targeted news to be recommended are based on matches with the current input content, because
This, the embodiment of the present invention can identify the current relevant News demand of input content exactly, and can be carried to active user
For the current relevant targeted news of input content, the accuracy that news is recommended can be also improved.
Also, the embodiment of the present invention recommend targeted news be it is closely related with the current input content of active user,
And the current input content for being typically different user is different, that is, having reflected active user in itself in current input content
Personalization it is special in the case of, the customized information that targeted news that the embodiment of the present invention is recommended can also agree with active user needs
It asks.
Description of the drawings
Fig. 1 is the schematic diagram that a kind of news of the present invention recommends the application environment of method;
Fig. 2 is that a kind of news of the present invention recommends the step flow chart of embodiment of the method one;
Fig. 3 is that a kind of news of the present invention recommends the step flow chart of embodiment of the method two;
Fig. 4 is a kind of structure diagram of news recommendation apparatus embodiment of the present invention;
Fig. 5 is a kind of block diagram of device 900 recommended for news of the present invention;And
Fig. 6 is the structure diagram of server in some embodiments of the present invention.
Specific embodiment
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, it is below in conjunction with the accompanying drawings and specific real
Applying mode, the present invention is described in further detail.
An embodiment of the present invention provides a kind of news suggested design, the program characterizes multiple users by news primary keyword
The hot news of concern, and the history input behavior data according to multiple users obtain the news primary keyword, in this way, current
When the current input content of user meets preset news correlated condition, illustrate the current input content be it is relevant with news,
Also it can determine that user deposits the relevant News demand of input content in this prior, therefore can obtain and recommend to user and deserve
The corresponding targeted news of news primary keyword that preceding input content matches;Since the embodiment of the present invention passes through preset news phase
Pass condition identifies user with the presence or absence of the current relevant News demand of input content only this is current exactly
News recommendation can be just carried out when input content meets preset news correlated condition, also, targeted news to be recommended are based on
News primary keyword match with the current input content, therefore, the embodiment of the present invention can identify exactly deserve
The preceding relevant News demand of input content, and the current relevant targeted news of input content can be provided to active user,
The accuracy of news recommendation can be improved.
Also, the embodiment of the present invention recommend targeted news be it is closely related with the current input content of active user,
And the current input content for being typically different user is different, that is, having reflected active user in itself in current input content
Personalization it is special in the case of, the customized information that targeted news that the embodiment of the present invention is recommended can also agree with active user needs
It asks.
In a kind of application example of the present invention, most users custom is being chatted during (passing through instant messaging program)
Input and the relevant content of news, such as:" Bryant is retired ", " you know media event B", " you know media event
The latest news of A ", " how many pieces of gold medals of the Olympic Games ", " lindane has match recently " etc., then the embodiment of the present invention can be from
It is excavated in the history input behavior data of multiple users:Characterize the news primary keyword (such as " section of the news of multiple user's concerns
Than ", " river teacher ", " Wang Baoqiang ", " Olympic Games ", " lindane " etc.);In this way, active user current input content meet it is preset
News correlated condition (for example, when the current input content and news primary keyword match), the main pass of the news can be obtained
The corresponding targeted news of keyword, and recommend active user.
The embodiment of the present invention can be applied to news website, news APP (application program, Application), input method
In the application environments such as APP, search APP, browser APP, to improve the accuracy of news recommendation;In addition, the embodiment of the present invention can
Recommended with being applied to top news, homepage, pop-up recommendation, input method Candidate Recommendation etc. in scene, it will be understood that the present invention is implemented
Example is not any limitation as specific application environment and specific recommendation scene.
News provided in an embodiment of the present invention recommends method to can be applied in application environment shown in FIG. 1, as shown in Figure 1,
Client 100 is located at server 200 in wired or wireless network, by the wired or wireless network, client 100 and service
Device 200 carries out data interaction.
Specifically, client 100 can obtain the current input content of active user, and send and deserve to server 200
Preceding input content;Alternatively, client 100 can obtain the current input content of active user, and to the current input content
After being analyzed, the analysis result of the current input content is sent to server 200;Optionally, it is above-mentioned in the current input
Appearance, which carries out analysis, to be included:Syntactic analysis is carried out to the current input content, therefrom to extract the vocabulary to play a crucial role,
Wherein, the vocabulary that auxiliary word etc. does not play a crucial role can be filtered out in extraction process, for example, the extraction from " Bryant is retired "
" Bryant ", for another example, from " you know media event B" in extraction " media event B " etc., it will be understood that the embodiment of the present invention
It is not any limitation as the detailed process analyzed the current input content.
Server 200 is judged after the current input content or corresponding analysis result is received in the current input
Hold or whether analysis result meets preset news correlated condition, if so, obtaining what is matched with the current input content
The corresponding targeted news of news primary keyword, and then export above-mentioned targeted news to client 100;
Above-mentioned targeted news can be shown to by client 100 after the reception transmission of server 200 is for targeted news
Active user checks.
Optionally, client 100 may operate on intelligent terminal, and above-mentioned intelligent terminal specifically includes but unlimited:Intelligence
Mobile phone, tablet computer, E-book reader, MP3 (dynamic image expert's compression standard audio level 3, Moving Picture
Experts Group Audio Layer III) player, MP4 (dynamic image expert's compression standard audio levels 4, Moving
Picture Experts Group Audio Layer IV) player, pocket computer on knee, vehicle-mounted computer, desk-top meter
Calculation machine, set-top box, intelligent TV set, wearable device etc..
Embodiment of the method one
With reference to Fig. 2, show that a kind of news of the present invention recommends the step flow chart of embodiment of the method one, can specifically wrap
Include following steps:
Step 201, the current input content for obtaining active user;
Step 202, when the current input content meets preset news correlated condition, obtain news primary keyword pair
The targeted news answered;Wherein, the news primary keyword matches with the current input content, and the news primary keyword is
History input behavior data according to multiple users obtain;
Step 203 recommends the targeted news to the active user.
The embodiment of the present invention can be applied to client or server;
Wherein, when applied to client, above-mentioned current input content for user by intelligent terminal in the application
The content of input, then client can obtain the corresponding targeted news of the current input content by performing step 202, and lead to
It crosses intelligent terminal and shows the targeted news to user;
When applied to server, above-mentioned current input content can be the content that client is sent, then server can be with
By performing step 202 to obtain the corresponding targeted news of above-mentioned current input content, and it is new to target described in client push
It hears.
The client of the embodiment of the present invention can be corresponding to the arbitrary APP run on intelligent terminal, wherein, which can be with
For input method APP, or news APP, browser APP or instant messaging APP etc.;Wherein, the clothes of input method APP
Business device can be collected and safeguard the history input behavior data of multiple users, and optionally, the server of input method APP can also carry
For the acquisition interface of the history input behavior data of the plurality of user so that other APP obtained by the acquisition interface it is the plurality of
The acquisition interface of the history input behavior data of user.The embodiment of the present invention is mainly illustrated by taking input method APP as an example, other
APP is cross-referenced.
It should be noted that the active user of the embodiment of the present invention can be marked by User ID (mark, Identity)
Know, can also be identified by the device id of intelligent terminal, be stepped on that is, the embodiment of the present invention is not required for user by User ID
Corresponding APP is recorded, different users can also be identified by device id.
In practical applications, when being inputted in application program of the active user in intelligent terminal, input method APP can be with
Corresponding current input content is captured in real time.News content is linked up by chatting in view of user's habituation, optionally,
The corresponding application environment of the current input content can be instant messaging program.It is appreciated that the embodiment of the present invention for
The corresponding application environment of the current input content is not any limitation as, for example, being related to user or the happiness of news for occupation
The joyous user for carrying out news analysis, can also input under the document application environment of such as OFFICE in relation to the current of news
Input content.
In the embodiment of the present invention, which can be one or more vocabulary, or one or
Multiple sentences.Optionally, under instant communication program environment, which can be the edit box of instant messaging program
In content to be sent, alternatively, the current input content can be instant messaging program chat window in transmission content
Reception content etc., that is, the current input content can include:Content to be sent has sent content and reception content
At least one of.Wherein, this sent content and reception content be for the opposite end of instant messaging, for example,
Active user chats with user A, then the opposite end of instant messaging is the intelligent terminal of user A.To sum up, in the embodiment of the present invention, when
The current input content of preceding user can not only include the content of active user's input, can also include being led to active user
The content of the peer user input of news, it will be understood that the embodiment of the present invention is not any limitation as specific current input content.
In the embodiment of the present invention, preset news correlated condition can be used for constraining the phase between current input content and news
Guan Xing.Optionally, since news primary keyword is obtained according to the history input behavior data of multiple users, can characterize
The news of multiple user's concerns, therefore above-mentioned current input content meets preset news correlated condition, can include:This is current defeated
Enter content with news primary keyword to match, that is, in the presence of the news primary keyword to match with the current input content.It is optional
Ground can store all news primary keyword by the list of news primary keyword, in this way, can by the current input content with
Each news primary keyword is matched in news primary keyword list.Optionally, the current input content and news primary keyword
Match, can specifically include:Vocabulary is same or similar or related to news primary keyword in the current input content,
That is, vocabulary and news primary keyword can be near synonym or related term in the current input content, the embodiment of the present invention for
The current input content matches with news primary keyword not to be any limitation as.In practical applications, user is for same news pair
There are " AlphaGo ", " Ah Fa dog " as that there can be different a variety of addresses, such as the address of " Ah method dog ", for " king is precious
Address by force " has " baby ", " stupid root " etc., and above-mentioned user can be used as corresponding for the multiple programs of same news object
Related term, it will be understood that the embodiment of the present invention is not any limitation as the particularly relevant word of news primary keyword.
Additionally, it is appreciated that vocabulary is same or similar or related to news primary keyword in the current input content
The alternative embodiment that the current input content matches with news primary keyword is intended only as, in fact, those skilled in the art
Can also be according to practical application request, other situations for being matched using the current input content and news primary keyword, for example,
The analysis result of the current input content and news primary keyword match etc..
It is related between current input content and news in order to accurately constrain in a kind of alternative embodiment of the present invention
Property, above-mentioned current input content meets preset news correlated condition, can include:The current input content and the main key of news
Word matches, and the current input content meets preset news statement pattern.Wherein, which can use
In constraining, news is corresponding to talk about sentence.Wherein, which can be according to multiple users or active user
History input behavior data obtain.In a kind of application example of the present invention, it is assumed that the embodiment of the present invention is from a large number of users
It user is excavated in history input behavior data talks about the pattern of news and be:" news for listening XXX ", " you know XXX ",
And assume that the embodiment of the present invention has excavated news primary keyword " Ah method dog " from the history input behavior data of multiple users,
In this way, once " you listened the news of Ah method dog for active user or the input of its peer user", then in corresponding current input
Hold not only comprising news keyword " Ah method dog ", but also meet preset news statement pattern " news for listening XXX ", therefore think this
Current input content meets preset news statement pattern.It is appreciated that those skilled in the art can be according to practical application need
It asks using required preset news statement pattern, the embodiment of the present invention is not subject to for specific preset news statement pattern
Limitation.
In the embodiment of the present invention, it is new that one or more can be excavated from the history input behavior data of multiple users
Hear primary keyword;Wherein, the plurality of user can be all or part of user in the whole network user, and input method APP can lead to
Cross the history input behavior data that client collects each user.Also, for the timeliness for ensureing news primary keyword, the history
Input behavior data can be the data of a preset time period, the initial time of the preset time period and the difference of current time
Can be less than time threshold, for example, the time threshold can be 2 it is small when, 12 it is small when, 24 it is small when, 48 it is small when or 72 it is small when
Deng it is possible thereby to ensure the timeliness of news primary keyword.Certainly, for the ease of analysis, more than the time threshold starting when
Between (for example, the history input behavior data before 100 days) also in protection scope of the present invention, the embodiment of the present invention is defeated for history
Enter the corresponding preset time period of behavioral data not to be any limitation as.
In a kind of alternative embodiment of the present invention, news primary keyword can be excavated as follows:To multiple use
The vocabulary attribute of each vocabulary is analyzed in the history input behavior data at family;Word is chosen from the history input behavior data
The attribute that converges meets the vocabulary of preset attribute condition, as news primary keyword.Optionally, above-mentioned vocabulary attribute includes:Word frequency and
At least one of number of users.Wherein, word frequency namely the occurrence number of certain vocabulary, optionally, the word frequency can be total word frequency,
Or the word frequency in chronomere (such as one day);Number of users namely certain vocabulary were inputted by how many user, similarly, the use
Amount amount can be total number of users, or the number of users in chronomere (such as one day).
In another alternative embodiment of the present invention, above-mentioned preset attribute condition can specifically include:
Increment rate of the word frequency in the first preset time period is more than first threshold;And/or
Increment rate of the number of users in the second preset time period is more than second threshold.
Wherein, increment rate of the word frequency in the first preset time period is more than that first threshold can correspond to the feelings that word frequency explodes
Shape.For example, certain vocabulary was at past 100 days, daily input quantity only has 50000 or so, and the input quantity of today explodes suddenly, becomes
100000, then it is related with this word namely may have occurred relevant new with the vocabulary to illustrate centainly to have occurred any thing
It hears.
Increment rate of the word frequency in the first preset time period is more than that first threshold can correspond to the situation that word frequency is stepped up.
For example, the word frequency of some vocabulary is incremented by by day in the first preset time period, day before yesterday 4w increased to 5w yesterdays, and became 6w etc. today,
It is considered that the vocabulary may be related to news.
Increment rate of the number of users in the second preset time period is more than that second threshold can correspond to the use for inputting certain vocabulary
The situation that amount amount explodes.For example, the original daily number of users of certain vocabulary is 5000, now daily number of users becomes suddenly
For 10000, then it is also assumed that the vocabulary may be related to news.
Increment rate of the above-mentioned word frequency in the first preset time period can be expressed as:Two days adjacent (one day after with the previous day)
Word frequency difference and the word frequency of the previous day ratio, increment rate of the above-mentioned number of users in the second preset time period can represent
For:The number of users difference of two days adjacent (one day after with the previous day) and the ratio of the number of users of the previous day, it will be understood that this
Field technology personnel can be according to practical application request, using rational first threshold and second threshold, the embodiment of the present invention pair
It is without restriction in specific first threshold and second threshold.
Additionally, it is appreciated that above-mentioned preset attribute condition is intended only as the alternative embodiment of the present invention, in fact, ability
Field technique personnel can also be according to practical application request, using other preset attribute conditions, as word frequency or number of users are more than
Corresponding threshold value etc., the embodiment of the present invention are not any limitation as specific preset attribute condition.
In practical applications, noise may be included in the numerous news primary keywords excavated, that is, some vocabulary can
It can be erroneously identified as news primary keyword, but actually itself and one news of non-corresponding.For above-mentioned noise problem, in this hair
In a kind of bright alternative embodiment, can also noise filtering be carried out to the news primary keyword by pseudo-linear filter mode.
Wherein, above-mentioned pseudo-linear filter mode can carry out news search using the news primary keyword as search term, and according to phase
The first news search result answered verifies whether the news primary keyword is noise.
The present invention another alternative embodiment in, it is described by pseudo-linear filter mode to the news primary keyword
The step of carrying out noise filtering, can specifically include:It searches for obtain corresponding first news according to the news primary keyword and search
Hitch fruit;The corresponding associative key of the news primary keyword is excavated from first news search result;According to described in
Associative key is searched for obtain corresponding second news search result;It is searched according to the news primary keyword in second news
Appearance information in hitch fruit judges whether the news primary keyword is noise, if so, filtering out the main key of the news
Word.It is alternatively possible to N number of first news search result for coming front is chosen, in the news of N number of first news search result
Appearance is segmented, and counts the word frequency of each vocabulary, therefrom chooses the highest M vocabulary of word frequency, as associative key, then will
The M associative key is scanned for as search term, to obtain corresponding second news search result, wherein, M, N are certainly
So number.Appearance information of the news primary keyword in the second news search result can include:Occurrence number, reverse document-frequency
Deng can then occur information according to this and judge whether the news primary keyword is noise.Corresponding judgment mode can there are many,
If for example, ratio of news primary keyword occurrence number and the number of the second news search result in the second news search result
More than the 3rd threshold value, it may be considered that the news primary keyword is not noise, therefore it can be retained, otherwise, can be somebody's turn to do
News primary keyword is noise.
It is appreciated that above-mentioned pseudo-linear filter mode of crossing is to the mode of news primary keyword progress noise filtering
As alternative embodiment, in fact, the mode of other noise filterings may be employed in those skilled in the art, for example, manual examination and verification
Mode or user's ballot mode etc., the embodiment of the present invention to the news primary keyword for carrying out noise filtering
Concrete mode be not any limitation as.
It should be noted that those skilled in the art can be determined according to time application demand first threshold, second threshold,
The concrete numerical value of 3rd threshold value, for example, first threshold, second threshold, the tool of the 3rd threshold value can be determined by way of experiment
Body numerical value, the embodiment of the present invention is for the concrete numerical value and its method of determination of the threshold values such as first threshold, second threshold, the 3rd threshold value
It is not any limitation as.
In the embodiment of the present invention, optionally, the step of the above-mentioned acquisition news primary keyword corresponding targeted news,
Can specifically it include:
The corresponding targeted news of the news primary keyword are searched in news database;Or
Recommend to obtain the corresponding target of the news primary keyword in news list from the news primary keyword is corresponding
News.
Wherein, the corresponding targeted news of the news primary keyword are searched in news database can correspond to on-line search
Mode, it is ensured that the timeliness of targeted news.Specifically, the query interface that can be provided by news database, with new
It hears primary keyword and carries out news search for search term, and can choose and come news P first, as targeted news.
Recommend to obtain the corresponding target of the news primary keyword in news list from the news primary keyword is corresponding
News can correspond to the mode of off-line search.That is, under offline mode, it can be beforehand through the inquiry of news database offer
Interface carries out news search, to obtain corresponding recommendation news list by search term of news primary keyword.Then step 202 can
With from the news primary keyword it is corresponding recommend to choose in news list come before the news of P, as targeted news, relatively
In the mode of on-line search, the mode of off-line search can improve the acquisition efficiency of targeted news.It is appreciated that targeted news
Quantity P can be determined by those skilled in the art according to practical application request.
In a kind of application example of the present invention, it is assumed that the history input behavior data based on a large number of users excavate " section
Than ", the news primary keyword such as " river teacher ", then the process chatted in active user by instant messaging environment, if input " section
It is more retired # " than #, it is possible to the corresponding several targeted news for recommending to search based on " Bryant " to active user;If input
" your # know # rivers teacher # media event # #" just new based on several targets that " river teacher " searches to active user's recommendation
It hears.Optionally, which can click directly on link and watch recommended targeted news, can also be directly by targeted news
It is shared with the peer user chatted.
The embodiment of the present invention can be applied to top news, homepage, pop-up recommendation, input method Candidate Recommendation etc. and recommend scene
In.For example, in the application environments such as news website, news APP, above-mentioned targeted news can be showed in its top position, specifically
Ground, certain user had just inputted content XXX, then for the user after news APP is opened, news APP can pass through pop-up mode
Show with the relevant targeted news of XXX, alternatively, it is more significant that top position etc. will be presented in the relevant targeted news of XXX
Position.For another example, in the application environment of input method APP, above-mentioned targeted news can be showed around its candidate window, for example,
If user just inputted " the retired # of Bryant # ", input method APP can show corresponding targeted news at the first time;Or
Person, input method APP can also show above-mentioned targeted news by pop-up mode.It is appreciated that the embodiment of the present invention for work as
Preceding user recommends the concrete mode of targeted news not to be any limitation as.
To sum up, the news of the embodiment of the present invention recommends method, and by preset news correlated condition, user whether there is
The current relevant News demand of input content is identified that only the current input content meets preset news exactly
News recommendation can be just carried out during correlated condition, also, the news primary keyword that targeted news to be recommended are based on is with deserving
What preceding input content matched, therefore, the embodiment of the present invention can identify that the current relevant news of input content needs exactly
It asks, and the current relevant targeted news of input content can be provided to active user, can also improve the standard of news recommendation
Exactness.
Also, the embodiment of the present invention recommend targeted news be it is closely related with the current input content of active user,
And the current input content for being typically different user is different, that is, having reflected active user in itself in current input content
Personalization it is special in the case of, the customized information that targeted news that the embodiment of the present invention is recommended can also agree with active user needs
It asks.
Embodiment of the method two
With reference to Fig. 3, show that a kind of news of the present invention recommends the step flow chart of embodiment of the method two, can specifically wrap
Include following steps:
Step 301, the current input content for obtaining active user;
Step 302, when the current input content meets preset news correlated condition, obtain news primary keyword and
The corresponding targeted news of associative key;Wherein, the news primary keyword matches with the current input content, described new
It hears primary keyword to obtain for the history input behavior data according to multiple users, the associative key and the main key of the news
Word is corresponding;
Step 303 recommends the targeted news to the active user.
Compared with embodiment of the method one shown in Fig. 2, the present embodiment can obtain news primary keyword and associative key pair
The targeted news answered, therefore more accurately news can be recommended to active user.
In the embodiment of the present invention, the current input content meets preset news correlated condition, can specifically include such as
Any in lower condition:
The current input content matches with news primary keyword;Or
The current input content matches with news primary keyword, and the current input content meets preset news
Statement pattern;Or
The current input content matches with news primary keyword, and the current input content and the main pass of the news
The corresponding associative key of keyword matches.
Wherein, the homogeneous matched condition of current input content and news primary keyword and associative key, can pass through phase
Close keyword be further ensured that on the basis of news primary keyword the current input content of active user be it is relevant with news,
Therefore more accurately news can be recommended to active user.For example, with a news primary keyword " river teacher " relevant news
Very much, once simultaneously containing associative key (such as the pass included in " river teacher " relevant news in the current input content of user
Keyword), then the scope for recommending news can be further reduced using news primary keyword+associative key as search term,
That is the locking relevant news of media event B recommends user, so as to ensure the precision recommended.
In a kind of application example of the present invention, it is assumed that the history input behavior data based on a large number of users excavate " section
Than ", the news primary keyword such as " river teacher ", and excavate " Bryant " corresponding associative key:" retired ", " match " etc. and
Excavate " river teacher " corresponding associative key, the then process chatted in active user by instant messaging environment, if defeated
Enter " the retired # of Bryant # ", it is possible to recommend " Bryant+retired ", " Bryant+match " if what is searched is corresponding to active user
Dry targeted news;If " your # know # rivers teacher # news # # for input" just recommend to active user based on " river teacher+related keyword
Several targeted news that word 1 ", " river teacher+associative key 2 " search.It is appreciated that the related keyword corresponding to targeted news
Word can be the associative key that current input content includes, alternatively, the associative key corresponding to targeted news can be to work as
The associative key that preceding input content does not include, therefore accurate and abundant news can be provided a user.
In a kind of alternative embodiment of invention, step excavates the side of the corresponding associative key of the news primary keyword
Formula can specifically include:
Mode 1 excavates the corresponding related pass of the news primary keyword from the history input behavior data of multiple users
Keyword;And/or
Mode 2 searches for obtain corresponding news search result according to the news primary keyword, from the news search knot
The corresponding associative key of the news primary keyword is excavated in fruit.
Wherein, mode 1 can analyze the history input behavior data of multiple users, for example, can count multiple
The vocabulary occurred simultaneously with news primary keyword in the history input behavior data of user, and choose same occurrence according to co-occurrence number
Number is more than the vocabulary of the 4th threshold value, as associative key.Optionally, the vocabulary of the 4th threshold value can also be more than to co-occurrence number
Further filtered.For example, being more than the vocabulary of the 4th threshold value for co-occurrence number, judge it whether in news primary keyword pair
Occur in the news search result answered, if so, as associative key, otherwise filter this out.It for another example, can will be same
Vocabulary+news keyword that occurrence number is more than the 4th threshold value carries out news search as search term, judges corresponding news search
As a result whether news search result corresponding to news keyword is related, if so, as associative key, otherwise by it
Filter out etc., it will be understood that the embodiment of the present invention is not any limitation as specific filter type.
Mode 2 can search for obtain corresponding first news search result according to the news primary keyword;And from described
The corresponding associative key of the news primary keyword is excavated in first news search result.It is alternatively possible to before choosing and coming
N number of first news search result in face, segments the news content of N number of first news search result, and counts each vocabulary
Word frequency, therefrom choose the highest M vocabulary of word frequency, as associative key.
It is appreciated that aforesaid way 1 and mode 2 are intended only as the optional digging of the corresponding associative key of news primary keyword
Pick mode, in fact, those skilled in the art can excavate mode, for example, manually according to practical application request using arbitrary
Excavation mode for another example, selects the target search word for including news primary keyword, then, by the mesh from hot news search term
It marks except the vocabulary between news primary keyword is as associative key etc. in search term, the embodiment of the present invention is for the main key of news
The specific excavation mode of the corresponding associative key of word is not any limitation as.
To sum up, the news of the embodiment of the present invention recommends method, has the following advantages that:
First, in the case of without manual intervention, according to the history input behavior data of multiple users, automatic mining goes out
The news primary keyword of the hot news of multiple user's concerns is characterized, different from traditional hot news search term, the present invention is real
The news primary keyword for applying example is derived from history input behavior data, although because some news are popular, user is not necessarily
Relevant vocabulary is inputted frequently by input method APP, the mode that the input quantity so based on vocabulary excavates hot news may be difficult
To excavate this kind of news, for example, during the input quantity based on vocabulary excavates hot news, input quantity row may be excavated
In vocabulary P first, as the keyword of hot news, and the vocabulary after P can not be excavated, wherein, P be more than or equal to
1 integer, such as P can be 10;And the embodiment of the present invention is accustomed to (passing through instant messaging journey in chat using most users
Sequence) input and the rule of the relevant content of news, the excavation of progress news primary keyword can obtain meeting most of in the process
The news primary keyword of the custom of user, and the coverage rate of the news primary keyword of characterization hot news can be improved;
Secondly, in the case of without manual intervention, the news primary keyword that is excavated by pseudo-linear filter mode
The noise data contained;
Furthermore the current input content based on active user, for recommending to meet the current input content from trend user
Property news.
In addition, by preset news correlated condition, user is needed with the presence or absence of the current relevant news of input content
It asks and is identified exactly, news can be just carried out when only the current input content meets preset news correlated condition and is pushed away
It recommends, also, the news primary keyword that targeted news to be recommended are based on matches with the current input content, therefore,
The embodiment of the present invention can identify the current relevant News demand of input content exactly, and can provide and be somebody's turn to do to active user
The relevant targeted news of current input content can also improve the accuracy of news recommendation.
Also, news primary keyword and the corresponding targeted news of associative key are obtained, therefore can be pushed away to active user
Recommend more accurately news.
It should be noted that for embodiment of the method, in order to be briefly described, therefore it is dynamic that it is all expressed as to a series of movement
It combines, but those skilled in the art should know, the embodiment of the present invention and from the limit of described athletic performance order
System, because according to the embodiment of the present invention, some steps may be employed other orders or be carried out at the same time.Secondly, art technology
Personnel should also know that embodiment described in this description belongs to preferred embodiment, and involved athletic performance simultaneously differs
Surely necessary to being the embodiment of the present invention.
Device embodiment
With reference to Fig. 4, show a kind of structure diagram of news recommendation apparatus embodiment of the present invention, can specifically include:
Input content acquisition module 401, news acquisition module 402 and news recommending module 403.
Wherein, input content acquisition module 401, for obtaining the current input content of active user;
News acquisition module 402 is new for when the current input content meets preset news correlated condition, obtaining
The corresponding targeted news of primary keyword are heard, alternatively, obtaining news primary keyword and the corresponding targeted news of associative key;Its
In, the news primary keyword matches with the current input content, and the news primary keyword is according to multiple users'
History input behavior data obtain, and the associative key is corresponding to the news primary keyword;
News recommending module 403, for recommending the targeted news to the active user.
Optionally, the current input content meets preset news correlated condition, can include:
The current input content matches with news primary keyword;Or
The current input content matches with news primary keyword, and the current input content meets preset news
Statement pattern;Or
The current input content matches with news primary keyword, and the current input content and the main pass of the news
The corresponding associative key of keyword matches.
Optionally, described device can also include:For excavating the first of news primary keyword the excavation module;
Described first excavates module, can include:
Submodule is analyzed, the vocabulary attribute for each vocabulary in the history input behavior data to multiple users divides
Analysis;And
Submodule is chosen, meets preset attribute condition for choosing vocabulary attribute from the history input behavior data
Vocabulary, as news primary keyword.
Optionally, the vocabulary attribute can include:At least one of word frequency and number of users.
Optionally, the preset attribute condition can include:
Increment rate of the word frequency in the first preset time period is more than first threshold;And/or
Increment rate of the number of users in the second preset time period is more than second threshold.
Optionally, described device can also include:
Noise filtering module, for carrying out noise filtering to the news primary keyword by pseudo-linear filter mode.
Optionally, the noise filtering module can include:
First news search submodule, for searching for obtain corresponding first news search according to the news primary keyword
As a result;
Associative key excavates submodule, for excavating the news primary keyword from first news search result
Corresponding associative key;
Second news search submodule, for searching for obtain corresponding second news search knot according to the associative key
Fruit;
Judging submodule, for the believing in second news search result according to the news primary keyword
Breath, judges whether the news primary keyword is noise, if so, filtering out the news primary keyword.
Optionally, described device can also include:For excavating the corresponding associative key of the news primary keyword
Second excavates module;
Described second excavates module, can include:
First excavates submodule, for excavating the news primary keyword from the history input behavior data of multiple users
Corresponding associative key;And/or
Second excavates submodule, for searching for obtain corresponding news search result according to the news primary keyword, from
The corresponding associative key of the news primary keyword is excavated in the news search result.
Optionally, the news acquisition module can include:
3rd news search submodule is new for searching for the corresponding target of the news primary keyword in news database
It hears;Or
List acquisition submodule, for recommending to obtain the news in news list from the news primary keyword is corresponding
The corresponding targeted news of primary keyword.
For device embodiment, since it is basicly similar to embodiment of the method, so description is fairly simple, it is related
Part illustrates referring to the part of embodiment of the method.
Each embodiment in this specification is described by the way of progressive, the highlights of each of the examples are with
The difference of other embodiment, just to refer each other for identical similar part between each embodiment.
On the device in above-described embodiment, wherein modules perform the concrete mode of operation in related this method
Embodiment in be described in detail, explanation will be not set forth in detail herein.
Fig. 5 is the block diagram according to a kind of device 900 recommended for news shown in an exemplary embodiment.For example, dress
It can be mobile phone to put 900, computer, digital broadcast terminal, messaging devices, game console, tablet device, medical treatment
Equipment, body-building equipment, personal digital assistant etc..
With reference to 5, device 900 can include following one or more assemblies:Processing component 902, memory 904, power supply group
Part 906, multimedia component 908, audio component 910, the interface 912 of input/output (I/O), sensor module 914, Yi Jitong
Believe component 916.
The integrated operation of 902 usual control device 900 of processing component, such as with display, call, data communication, phase
Machine operates and record operates associated operation.Processing element 902 can refer to including one or more processors 920 to perform
Order, to perform all or part of the steps of the methods described above.In addition, processing component 902 can include one or more modules, just
Interaction between processing component 902 and other assemblies.For example, processing component 902 can include multi-media module, it is more to facilitate
Interaction between media component 908 and processing component 902.
Memory 904 is configured as storing various types of data to support the operation in equipment 900.These data are shown
Example is included for the instruction of any application program or method that are operated on device 900, contact data, and telephone book data disappears
Breath, picture, video etc..Memory 904 can be by any kind of volatibility or non-volatile memory device or their group
Close and realize, such as static RAM (SRAM), electrically erasable programmable read-only memory (EEPROM), it is erasable can
Program read-only memory (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, flash memory
Reservoir, disk or CD.
Power supply module 906 provides electric power for the various assemblies of device 900.Power supply module 906 can include power management system
System, one or more power supplys and other generate, manage and distribute electric power associated component with for device 900.
Multimedia component 908 is included in the screen of one output interface of offer between described device 900 and user.One
In a little embodiments, screen can include liquid crystal display (LCD) and touch panel (TP).If screen includes touch panel, screen
Curtain may be implemented as touch-screen, to receive input signal from the user.Touch panel includes one or more touch sensings
Device is to sense the gesture on touch, slide, and touch panel.The touch sensor can not only sense touch or sliding motion
The border of action, but also detect duration and pressure associated with the touch or slide operation.In some embodiments,
Multimedia component 908 includes a front camera and/or rear camera.When equipment 900 is in operation mode, mould is such as shot
When formula or video mode, front camera and/or rear camera can receive external multi-medium data.Each preposition camera shooting
Head and rear camera can be a fixed optical lens system or have focusing and optical zoom capabilities.
Audio component 910 is configured as output and/or input audio signal.For example, audio component 910 includes a wheat
Gram wind (MIC), when device 900 is in operation mode, during such as call model, logging mode and speech recognition mode, microphone quilt
It is configured to receive external audio signal.The received audio signal can be further stored in memory 904 or via communication
Component 916 is sent.In some embodiments, audio component 910 further includes a loud speaker, for exports audio signal.
I/O interfaces 912 provide interface between processing component 902 and peripheral interface module, and above-mentioned peripheral interface module can
To be keyboard, click wheel, button etc..These buttons may include but be not limited to:Home button, volume button, start button and lock
Determine button.
Sensor module 914 includes one or more sensors, and the state for providing various aspects for device 900 is commented
Estimate.For example, sensor module 914 can detect opening/closed state of equipment 900, and the relative positioning of component, for example, it is described
Component is the display and keypad of device 900, and sensor module 914 can be with 900 1 groups of detection device 900 or device
The position change of part, the existence or non-existence that user contacts with device 900,900 orientation of device or acceleration/deceleration and device 900
Temperature change.Sensor module 914 can include proximity sensor, be configured to examine without any physical contact
Survey presence of nearby objects.Sensor module 914 can also include optical sensor, such as CMOS or CCD imaging sensors, be used for
It is used in imaging applications.In some embodiments, which can also include acceleration transducer, gyroscope
Sensor, Magnetic Sensor, pressure sensor or temperature sensor.
Communication component 916 is configured to facilitate the communication of wired or wireless way between device 900 and other equipment.Device
900 can access the wireless network based on communication standard, such as WiFi, 2G or 3G or combination thereof.In an exemplary implementation
In example, communication component 916 receives broadcast singal or broadcast related information from external broadcasting management system via broadcast channel.
In one exemplary embodiment, the communication component 916 further includes near-field communication (NFC) module, to promote short range communication.Example
Such as, NFC module can be based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra wide band (UWB) technology,
Bluetooth (BT) technology and other technologies are realized.
In the exemplary embodiment, device 900 can be by one or more application application-specific integrated circuit (ASIC), number
Signal processor (DSP), digital signal processing appts (DSPD), programmable logic device (PLD), field programmable gate array
(FPGA), controller, microcontroller, microprocessor or other electronic components are realized, for performing the above method.
In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium including instructing, example are additionally provided
Such as include the memory 904 of instruction, above-metioned instruction can be performed to complete the above method by the processor 920 of device 900.For example,
The non-transitorycomputer readable storage medium can be ROM, random access memory (RAM), CD-ROM, tape, floppy disk
With optical data storage devices etc..
A kind of non-transitorycomputer readable storage medium, when the instruction in the storage medium is by the processing of intelligent terminal
When device performs so that intelligent terminal is able to carry out a kind of news and recommends method, the described method includes:Obtain the current of active user
Input content;When the current input content meets preset news correlated condition, the corresponding mesh of news primary keyword is obtained
News is marked, alternatively, obtaining news primary keyword and the corresponding targeted news of associative key;Wherein, the news primary keyword
Match with the current input content, the news primary keyword obtains for the history input behavior data according to multiple users
It arrives, the associative key is corresponding to the news primary keyword;Recommend the targeted news to the active user.
Fig. 6 is the structure diagram of server in some embodiments of the present invention.The server 1900 can be because of configuration or property
Energy is different and generates bigger difference, can include one or more central processing units (central processing
Units, CPU) 1922 (for example, one or more processors) and memory 1932, one or more storage applications
The storage medium 1930 of program 1942 or data 1944 (such as one or more mass memory units).Wherein, memory
1932 and storage medium 1930 can be of short duration storage or persistent storage.Being stored in the program of storage medium 1930 can include
One or more modules (diagram does not mark), each module can include operating the series of instructions in server.More
Further, central processing unit 1922 could be provided as communicating with storage medium 1930, and storage is performed on server 1900 and is situated between
Series of instructions operation in matter 1930.
Server 1900 can also include one or more power supplys 1926, one or more wired or wireless nets
Network interface 1950, one or more input/output interfaces 1958, one or more keyboards 1956 and/or, one or
More than one operating system 1941, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM
Etc..
Those skilled in the art will readily occur to the present invention its after considering specification and putting into practice invention disclosed herein
Its embodiment.It is contemplated that cover the present invention any variations, uses, or adaptations, these modifications, purposes or
Person's adaptive change follows the general principle of the present invention and including the undocumented common knowledge in the art of the disclosure
Or conventional techniques.Description and embodiments are considered only as illustratively, and true scope and spirit of the invention are by following
Claim is pointed out.
It should be appreciated that the invention is not limited in the precision architecture for being described above and being shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.The scope of the present invention is only limited by appended claim
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all the present invention spirit and
Within principle, any modifications, equivalent replacements and improvements are made should all be included in the protection scope of the present invention.
Method, a kind of news recommendation apparatus and one kind is recommended to be pushed away for news a kind of news provided by the present invention above
The device recommended, is described in detail, and specific case used herein explains the principle of the present invention and embodiment
It states, the explanation of above example is only intended to help to understand method and its core concept of the invention;Meanwhile for this field
Those skilled in the art, thought according to the invention, in specific embodiments and applications there will be changes, to sum up institute
It states, this specification content should not be construed as limiting the invention.
Claims (11)
1. a kind of news recommends method, which is characterized in that including:
Obtain the current input content of active user;
When the current input content meets preset news correlated condition, it is new to obtain the corresponding target of news primary keyword
It hears, alternatively, obtaining news primary keyword and the corresponding targeted news of associative key;Wherein, the news primary keyword and institute
It states current input content to match, the news primary keyword obtains for the history input behavior data according to multiple users, institute
It is corresponding to the news primary keyword to state associative key;
Recommend the targeted news to the active user.
2. according to the method described in claim 1, it is characterized in that, the current input content meets preset news correlation item
Part, including:
The current input content matches with news primary keyword;Or
The current input content matches with news primary keyword, and the current input content meets preset news sentence
Pattern;Or
The current input content matches with news primary keyword, and the current input content and the news primary keyword
Corresponding associative key matches.
3. according to the method described in claim 1, it is characterized in that, news primary keyword is excavated as follows:
The vocabulary attribute of each vocabulary in the history input behavior data of multiple users is analyzed;
The vocabulary that vocabulary attribute meets preset attribute condition is chosen from the history input behavior data, as the main key of news
Word.
4. according to the method described in claim 3, it is characterized in that, the vocabulary attribute includes:In word frequency and number of users
It is at least one.
5. according to the method described in claim 4, it is characterized in that, the preset attribute condition includes:
Increment rate of the word frequency in the first preset time period is more than first threshold;And/or
Increment rate of the number of users in the second preset time period is more than second threshold.
6. according to any method in claim 3 to 5, which is characterized in that the method further includes:
Noise filtering is carried out to the news primary keyword by pseudo-linear filter mode.
7. according to the method described in claim 6, it is characterized in that, it is described by pseudo-linear filter mode to the main pass of the news
Keyword carries out the step of noise filtering, including:
It searches for obtain corresponding first news search result according to the news primary keyword;
The corresponding associative key of the news primary keyword is excavated from first news search result;
It searches for obtain corresponding second news search result according to the associative key;
According to appearance information of the news primary keyword in second news search result, the main key of the news is judged
Whether word is noise, if so, filtering out the news primary keyword.
8. according to any method in claim 1 to 5, which is characterized in that excavate the news master as follows
The corresponding associative key of keyword:
The corresponding associative key of the news primary keyword is excavated from the history input behavior data of multiple users;And/or
It searches for obtain corresponding news search result according to the news primary keyword, institute is excavated from the news search result
State the corresponding associative key of news primary keyword.
9. according to any method in claim 1 to 5, which is characterized in that described to obtain the news primary keyword pair
The step of targeted news answered, including:
The corresponding targeted news of the news primary keyword are searched in news database;Or
Recommend to obtain the corresponding targeted news of the news primary keyword in news list from the news primary keyword is corresponding.
10. a kind of news recommendation apparatus, which is characterized in that including:
Input content acquisition module, for obtaining the current input content of active user;
News acquisition module, for when the current input content meets preset news correlated condition, obtaining the main pass of news
The corresponding targeted news of keyword, alternatively, obtaining news primary keyword and the corresponding targeted news of associative key;Wherein, it is described
News primary keyword matches with the current input content, and the news primary keyword is the history input according to multiple users
Behavioral data obtains, and the associative key is corresponding to the news primary keyword;And
News recommending module, for recommending the targeted news to the active user.
11. a kind of device recommended for news, which is characterized in that include memory and one or more than one
Program, either more than one program storage in memory and is configured to by one or more than one processing for one of them
Device performs the one or more programs and includes the instruction operated below:
Obtain the current input content of active user;
When the current input content meets preset news correlated condition, it is new to obtain the corresponding target of news primary keyword
It hears, alternatively, obtaining news primary keyword and the corresponding targeted news of associative key;Wherein, the news primary keyword and institute
It states current input content to match, the news primary keyword obtains for the history input behavior data according to multiple users, institute
It is corresponding to the news primary keyword to state associative key;
Recommend the targeted news to the active user.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610995502.1A CN108073606B (en) | 2016-11-10 | 2016-11-10 | News recommendation method and device for news recommendation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610995502.1A CN108073606B (en) | 2016-11-10 | 2016-11-10 | News recommendation method and device for news recommendation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108073606A true CN108073606A (en) | 2018-05-25 |
CN108073606B CN108073606B (en) | 2021-12-28 |
Family
ID=62154721
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610995502.1A Active CN108073606B (en) | 2016-11-10 | 2016-11-10 | News recommendation method and device for news recommendation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108073606B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109145206A (en) * | 2018-07-31 | 2019-01-04 | 优视科技新加坡有限公司 | A kind of method, apparatus and equipment/terminal/server that news is shared |
CN110245243A (en) * | 2019-06-20 | 2019-09-17 | 北京百度网讯科技有限公司 | The method and apparatus of news retrieval, electronic equipment, computer-readable medium |
CN110598098A (en) * | 2019-08-30 | 2019-12-20 | 北京搜狗科技发展有限公司 | Information recommendation method and device and information recommendation device |
CN111222040A (en) * | 2019-12-30 | 2020-06-02 | 航天信息股份有限公司企业服务分公司 | Scheme self-matching processing method and system based on training requirement |
CN111291265A (en) * | 2020-02-10 | 2020-06-16 | 青岛聚看云科技有限公司 | Recommendation information generation method and device |
CN112328861A (en) * | 2020-11-24 | 2021-02-05 | 郑州航空工业管理学院 | News dissemination method based on big data processing |
CN112667894A (en) * | 2020-12-25 | 2021-04-16 | 特赞(上海)信息科技有限公司 | Content recommendation method, device, equipment and storage medium |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101431485A (en) * | 2008-12-31 | 2009-05-13 | 深圳市迅雷网络技术有限公司 | Method and system for automatically recommending internet information |
CN101446959A (en) * | 2008-12-30 | 2009-06-03 | 深圳市迅雷网络技术有限公司 | Internet-based news recommendation method and system thereof |
CN103399891A (en) * | 2013-07-22 | 2013-11-20 | 百度在线网络技术(北京)有限公司 | Method, device and system for automatic recommendation of network content |
CN103559265A (en) * | 2013-11-04 | 2014-02-05 | 北京中搜网络技术股份有限公司 | Individualized push method of cell phone client |
CN104036038A (en) * | 2014-06-30 | 2014-09-10 | 北京奇虎科技有限公司 | News recommendation method and system |
US20150074131A1 (en) * | 2013-09-09 | 2015-03-12 | Mobitv, Inc. | Leveraging social trends to identify relevant content |
CN104657393A (en) * | 2013-11-25 | 2015-05-27 | 深圳市至高通信技术发展有限公司 | Public opinion analysis method and corresponding device |
US20150169563A1 (en) * | 2013-12-13 | 2015-06-18 | National Cheng Kung University | Method and system for recommending research information news |
-
2016
- 2016-11-10 CN CN201610995502.1A patent/CN108073606B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101446959A (en) * | 2008-12-30 | 2009-06-03 | 深圳市迅雷网络技术有限公司 | Internet-based news recommendation method and system thereof |
CN101431485A (en) * | 2008-12-31 | 2009-05-13 | 深圳市迅雷网络技术有限公司 | Method and system for automatically recommending internet information |
CN103399891A (en) * | 2013-07-22 | 2013-11-20 | 百度在线网络技术(北京)有限公司 | Method, device and system for automatic recommendation of network content |
US20150074131A1 (en) * | 2013-09-09 | 2015-03-12 | Mobitv, Inc. | Leveraging social trends to identify relevant content |
CN103559265A (en) * | 2013-11-04 | 2014-02-05 | 北京中搜网络技术股份有限公司 | Individualized push method of cell phone client |
CN104657393A (en) * | 2013-11-25 | 2015-05-27 | 深圳市至高通信技术发展有限公司 | Public opinion analysis method and corresponding device |
US20150169563A1 (en) * | 2013-12-13 | 2015-06-18 | National Cheng Kung University | Method and system for recommending research information news |
CN104036038A (en) * | 2014-06-30 | 2014-09-10 | 北京奇虎科技有限公司 | News recommendation method and system |
Non-Patent Citations (2)
Title |
---|
文鹏;蔡瑞;吴黎兵: "一种基于潜在类别模型的新闻推荐方法", 《情报杂志》 * |
林中明; 李文敬: "基于Hadoop的Web用户识别与新闻智能推荐算法研究", 《软件导刊》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109145206A (en) * | 2018-07-31 | 2019-01-04 | 优视科技新加坡有限公司 | A kind of method, apparatus and equipment/terminal/server that news is shared |
CN110245243A (en) * | 2019-06-20 | 2019-09-17 | 北京百度网讯科技有限公司 | The method and apparatus of news retrieval, electronic equipment, computer-readable medium |
CN110598098A (en) * | 2019-08-30 | 2019-12-20 | 北京搜狗科技发展有限公司 | Information recommendation method and device and information recommendation device |
CN111222040A (en) * | 2019-12-30 | 2020-06-02 | 航天信息股份有限公司企业服务分公司 | Scheme self-matching processing method and system based on training requirement |
CN111222040B (en) * | 2019-12-30 | 2023-06-13 | 航天信息股份有限公司企业服务分公司 | Scheme self-matching processing method and system based on training requirements |
CN111291265A (en) * | 2020-02-10 | 2020-06-16 | 青岛聚看云科技有限公司 | Recommendation information generation method and device |
CN111291265B (en) * | 2020-02-10 | 2023-10-03 | 青岛聚看云科技有限公司 | Recommendation information generation method and device |
CN112328861A (en) * | 2020-11-24 | 2021-02-05 | 郑州航空工业管理学院 | News dissemination method based on big data processing |
CN112328861B (en) * | 2020-11-24 | 2023-06-23 | 郑州航空工业管理学院 | News spreading method based on big data processing |
CN112667894A (en) * | 2020-12-25 | 2021-04-16 | 特赞(上海)信息科技有限公司 | Content recommendation method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN108073606B (en) | 2021-12-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108073606A (en) | A kind of news recommends method and apparatus, a kind of device recommended for news | |
CN105488154A (en) | Theme application recommendation method and device | |
CN110175223A (en) | A kind of method and device that problem of implementation generates | |
CN109614482A (en) | Processing method, device, electronic equipment and the storage medium of label | |
CN104281703B (en) | The method and device of similarity calculation between uniform resource position mark URL | |
CN108121736A (en) | A kind of descriptor determines the method for building up, device and electronic equipment of model | |
CN107291772B (en) | Search access method and device and electronic equipment | |
CN108227950A (en) | A kind of input method and device | |
CN108874939A (en) | A kind of information search method and device | |
CN105373580A (en) | Method and device for displaying subjects | |
CN107918496A (en) | It is a kind of to input error correction method and device, a kind of device for being used to input error correction | |
CN104111979A (en) | Search recommendation method and device | |
CN106202150A (en) | Method for information display and device | |
CN112784142A (en) | Information recommendation method and device | |
CN108345625A (en) | A kind of information mining method and device, a kind of device for information excavating | |
CN110222256A (en) | A kind of information recommendation method, device and the device for information recommendation | |
CN109634913A (en) | Storage method, device and the electronic equipment of document | |
CN107045541A (en) | data display method and device | |
CN110019885A (en) | A kind of expression data recommended method and device | |
CN107402767A (en) | The method and apparatus for showing PUSH message | |
CN104636476A (en) | Friend recommending method and friend recommending device | |
CN110020309A (en) | A kind of page processing method and device | |
CN103970831B (en) | Recommend the method and apparatus of icon | |
CN110286775A (en) | A kind of dictionary management method and device | |
CN109783244A (en) | Treating method and apparatus, the device for processing |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |