CN110532478A - A kind of dissemination of news method based on big data processing - Google Patents
A kind of dissemination of news method based on big data processing Download PDFInfo
- Publication number
- CN110532478A CN110532478A CN201910833901.1A CN201910833901A CN110532478A CN 110532478 A CN110532478 A CN 110532478A CN 201910833901 A CN201910833901 A CN 201910833901A CN 110532478 A CN110532478 A CN 110532478A
- Authority
- CN
- China
- Prior art keywords
- news
- keyword
- user
- preference
- pushed
- 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
- G06F16/90—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
In order to solve problems in the prior art, present disclose provides a kind of dissemination of news methods based on big data processing, news are pushed to user based on keyword, to improve news push efficiency.Include: to obtain news to be pushed and its keyword, forms news database to be pushed;Wait news push trigger signal;If receiving news push trigger signal, keyword coefficient of relationship and user's news preference data are obtained;And it is based on user's news preference data and keyword coefficient of relationship, news to be pushed is obtained as user preference news;News to be pushed is obtained at random from news library to be pushed as user and tests news;The first news collection is pushed to user terminal, and obtains the feedback information of user terminal;Feedback information based on user terminal adjusts the keyword coefficient of relationship between keyword.The disclosure is based on big data and optimizes and revises keyword coefficient of relationship, and the news based on keyword coefficient of relationship push user preferences, improves news push.
Description
Technical field
This disclosure relates to a kind of dissemination of news method more particularly to a kind of dissemination of news method based on big data processing.
Background technique
Today's society, News genre is more and more, and daily news new increment is very huge, and user is without words when limited
Interior all news of browsing, the news push how to pay close attention to user is the emphasis studied now to user;The prior art
It is that news is classified by type by the way of, news push is carried out according to the type that user likes;Its deficiency is: new
It is limited to hear type, big with type news quantity, if carrying out news push by News genre, being pushed news is exactly user's happiness
Joyous news probability is lower;After generally requiring push 10 or more news, or even up to a hundred news of push, it just will appear a use
The desired news in family, news push low efficiency.
Summary of the invention
At least one of in order to solve the above-mentioned technical problem, present disclose provides a kind of news based on big data processing
Transmission method pushes news to user based on keyword, to improve news push efficiency.
A kind of dissemination of news method based on big data processing of the disclosure, is included in server end and executes following steps:
News to be pushed and its keyword are obtained, news data to be pushed is formed according to news to be pushed and its keyword
Library;
Wait news push trigger signal;
If receiving news push trigger signal, the keyword coefficient of relationship between keyword and use to be pushed are obtained
User's news preference data at family, user's news preference data include User ID and user preference keyword;
Based on user's news preference data and keyword coefficient of relationship, preset from wait push acquisition first in news database
The news to be pushed of record is as user preference news;
The news to be pushed for obtaining the second default record at random from news library to be pushed tests news as user;
The first news collection is pushed to user terminal, and obtains the feedback information of user terminal, wherein the first news collection
Test news including user preference news and user, the feedback information of user terminal include each news for concentrating of the first news whether
The information browsed;
Feedback information based on user terminal adjusts the keyword coefficient of relationship between keyword.
Optionally, method further includes executing following steps in server end:
By the keyword coefficient of relationship N between keywordijIt is initialized as follows,
In above formula, NijIndicate coefficient of relationship of i-th of keyword relative to j-th of keyword;When i=j, i-th crucial
Word is 1 relative to the coefficient of relationship of j-th of keyword;When i ≠ j, relationship system of i-th of keyword relative to j-th of keyword
Number isWherein, YijIndicate relation value of i-th of keyword relative to j-th of keyword, n is keyword total quantity, YijJust
Initial value is 1.
Optionally, method further includes executing following steps in user terminal:
Initialize the preference value of each keyword;
According to the summation of the preference value of each keyword and the preference value of all keywords, the preference value ratio of each keyword is obtained
Weight;
The preference value specific gravity of each keyword is arranged from big to small, obtains pass of the preference value than the first preset value before permutatation
The user preference keyword as the user of keyword;
Receive the second news of simultaneously display server end push;
Judge whether the second news is browsed;
If the second news is browsed, the preference value which corresponds to keyword increases by 1%, and otherwise, this is second new
The preference value for hearing corresponding keyword reduces 1%;
The preference value specific gravity of each keyword is arranged from big to small, obtains pass of the preference value than the first preset value before permutatation
The user preference keyword as the user of keyword.
Optionally, method further includes executing following steps in server end:
The preference value of each keyword of initialising subscriber;
According to the summation of the preference value of each keyword of user and the preference value of all keywords of user, obtain user's
The preference value specific gravity of each keyword;
The preference value specific gravity of each keyword of user is arranged from big to small, it is more default than before permutatation first to obtain preference value
The user preference keyword as the user of the keyword of value;
After the feedback information for obtaining user terminal;
Whether each news for judging that the first news is concentrated is browsed, and is increased by the preference value of the corresponding keyword of browsing news
1%, the preference value for the keyword not browsed reduces 1%;
The preference value specific gravity of each keyword of user is arranged from big to small, obtains preference value than the first preset value before permutatation
Keyword the user preference keyword as the user.
Optionally, it is based on user's news preference data and keyword coefficient of relationship, is obtained from wait push in news database
The news of first default record is as user preference news, comprising:
News to be pushed is obtained in news database from wait push, and extracts the keyword of respectively news to be pushed;
It is closed according to user's news that user preference keyword and the respectively keyword of news to be pushed calculate respectively news to be pushed
Join weight;
According to user's news associated weights of respectively news to be pushed, user's news associated weights are sorted from large to small, are obtained
Take the news to be pushed of the preceding first default record as user preference news.
Optionally, according to user preference keyword and respectively, the keyword of news to be pushed calculates the user of respectively news to be pushed
News associated weights include:
Obtain each keyword of news to be pushed and the coefficient of relationship of user preference keyword;
Each keyword for news to be pushed of summing and the coefficient of relationship of user preference keyword, obtain the use of news to be pushed
Family news associated weights.
Optionally, based on the feedback information of user terminal, the keyword coefficient of relationship adjusted between keyword includes:
Judge each news of the first news concentration whether clear by ustomer premises access equipment according to the feedback information of user terminal
It lookes at,
If the news that the first news is concentrated is browsed by ustomer premises access equipment, and the keyword by browsing news is opposite
Increase by 0.1% in the relation value of user preference keyword, and judge the relation value, if, will if relation value is more than n more than n
The relation value is set to 1;
If the news that the first news is concentrated is not browsed by ustomer premises access equipment, by this not by the keyword phase of browsing news
Reduction 0.1% for the relation value of user preference keyword.
Optionally, user's news preference data includes the preference value specific gravity of user preference keyword;
Feedback information based on user terminal, the keyword coefficient of relationship adjusted between keyword include:
Judge each news of the first news concentration whether clear by ustomer premises access equipment according to the feedback information of user terminal
It lookes at,
If the news that the first news is concentrated is browsed by ustomer premises access equipment, and the keyword by browsing news is opposite
Increase third preset value in the relation value of user preference keyword, and judge the relation value, if more than n, if relation value is more than
The relation value is then set to 1 by n, and the third preset value is the 0.1% of the preference value specific gravity of user preference keyword;
If the news that the first news is concentrated is not browsed by ustomer premises access equipment, by this not by the keyword phase of browsing news
For the 4th preset value of reduction of the relation value of user preference keyword, the 4th preset value is the inclined of user preference keyword
It is worth the 0.1% of specific gravity well.
Optionally, the first default record and the second default record equal 5.
Optionally, the first preset value is 5.
The utility model has the advantages that in disclosed method, the keyword coefficient of relationship between keyword passes through big one of the disclosure
The feedback data of the different user terminals of amount, iterates, the keyword coefficient of relationship between optimization keyword, and according to
Keyword coefficient of relationship between keyword after optimization carries out news push, so realize handled based on user's big data it is new
It hears and propagates, News Resources can be efficiently used and accurately pushed, improve news utilization rate and improve news push efficiency.
Detailed description of the invention
Attached drawing shows the illustrative embodiments of the disclosure, and it is bright together for explaining the principles of this disclosure,
Which includes these attached drawings to provide further understanding of the disclosure, and attached drawing is included in the description and constitutes this
Part of specification.
Fig. 1 is the running environment of the illustrative embodiments of the disclosure;
Fig. 2 is the method flow diagram of the illustrative embodiments of the disclosure.
Specific embodiment
The disclosure is described in further detail with embodiment with reference to the accompanying drawing.It is understood that this place
The specific embodiment of description is only used for explaining related content, rather than the restriction to the disclosure.It also should be noted that being
Convenient for description, part relevant to the disclosure is illustrated only in attached drawing.
It should be noted that in the absence of conflict, the feature in embodiment and embodiment in the disclosure can
To be combined with each other.The disclosure is described in detail below with reference to the accompanying drawings and in conjunction with embodiment.
Dissemination of news method provided in an embodiment of the present invention can be used for environment shown in FIG. 1, including user terminal and service
Device end, server end and user terminal are connected by network communication, wherein user terminal can be the equipment such as computer, mobile phone.Clothes
Business device end can be cloud server.
As shown in Fig. 2, a kind of dissemination of news method based on big data processing, for accurately pushing news to user, with
The propagation for realizing news is included in server end and executes following steps:
Step S1: obtaining news to be pushed and its keyword, is formed according to news to be pushed and its keyword new wait push
Hear database;
Step S2: news push trigger signal is waited;
Step S3: if receiving news push trigger signal, obtain keyword coefficient of relationship between keyword and
User's news preference data of user to be pushed, user's news preference data include User ID and user preference keyword;
Step S4: it is based on user's news preference data and keyword coefficient of relationship, is obtained from wait push in news database
The news to be pushed of first default record is as user preference news;
Step S5: the news to be pushed for obtaining the second default record at random from news library to be pushed is new as user's test
It hears;
Step S6: the first news collection is pushed to user terminal, and obtains the feedback information of user terminal, wherein described the
One news collection includes that user preference news and user test news;
Step S7: the feedback information based on user terminal adjusts the keyword coefficient of relationship between keyword.
News push trigger signal can according to need setting, is such as spaced preset time and automatically generates news push triggering letter
Number, for another example artificial push generates news push trigger signal.
Method in the present embodiment according between keyword keyword coefficient of relationship and user preference keyword to
Family pushes the first news collection (including user preference news and user test news), and according to the feedback information of user terminal, adjusts
Keyword coefficient of relationship between whole keyword optimizes the keyword coefficient of relationship between keyword, and according to the pass after optimization
Keyword coefficient of relationship between keyword carries out news push.
To sum up, the keyword coefficient of relationship between keyword is carried out by the feedback data of a large amount of different user terminals
It iterates, optimizes the keyword coefficient of relationship between keyword, and according to the keyword relationship between the keyword after optimization
Coefficient carries out news push, and then realizes the dissemination of news handled based on user's big data, and then can efficiently use news money
Source is accurately pushed, and news utilization rate is improved
In the step S1 of the present embodiment, the mode for obtaining push news and its keyword can be to record from newspapering personnel
It is obtained in the information entered, or by system grabs, can also be obtained using other means.
First default record and the second default record can according to need setting, and such as the first default record is set as 5.The
Two default records are set as 5.
Keyword coefficient of relationship between keyword is for indicating keyword relevance, in order to make it easy to understand, such as 1 institute of table
Show, the present embodiment indicates that the keyword coefficient of relationship between keyword, the i-th row j column indicate i-th of pass by the table of i row j column
Coefficient of relationship of the keyword relative to j-th of keyword;
Before step S1, further include the steps that initializing the keyword coefficient of relationship between keyword: will be between keyword
Keyword coefficient of relationship NijIt is initialized as follows,
In above formula, NijIndicate coefficient of relationship of i-th of keyword relative to j-th of keyword;When i=j, i-th crucial
Word is 1 relative to the coefficient of relationship of j-th of keyword;When i ≠ j, relationship system of i-th of keyword relative to j-th of keyword
Number isWherein, YijIndicate relation value of i-th of keyword relative to j-th of keyword, n is keyword total quantity, YijJust
Initial value is 1.
If there is new keyword, then by the keyword and other keywords during the method for executing the present embodiment
Coefficient of relationship be set to 1/n, after increasing new keywords, n can accordingly adjust;Such as: increase 1 keyword newly, then n increases by 1.
Of course, it will be appreciated that, YijInitial value may be the preset value of designer's input.
User 1 | Keyword 1 | Keyword 2 | Keyword 3 | …… | Keyword j |
Keyword 1 | N11 | N12 | N13 | …… | N1j |
Keyword 2 | N21 | N22 | N23 | …… | N2j |
Keyword 3 | N31 | N32 | N33 | …… | N3j |
…… | …… | …… | …… | …… | …… |
Keyword i | Ni1 | Ni2 | Ni3 | …… | Nij |
In step s 2, user's news preference data is obtained, user's news preference data includes User ID and user preference
Keyword;User preference keyword can be generated in server end, can also be generated in user terminal.
As optional, method further includes that user terminal executes following steps:
Initialize the preference value of each keyword;
According to the summation of the preference value of each keyword and the preference value of all keywords, the preference value ratio of each keyword is obtained
Weight;
The preference value specific gravity of each keyword is arranged from big to small, obtains pass of the preference value than the first preset value before permutatation
The user preference keyword as the user of keyword;
Receive and show the second news of push server end push;
Judge whether the second news is browsed;
If the second news is browsed, the preference value which corresponds to keyword increases by 1%, and otherwise, this is second new
The preference value for hearing corresponding keyword reduces 1%;
The preference value specific gravity of each keyword is arranged from big to small, obtains pass of the preference value than the first preset value before permutatation
The user preference keyword as the user of keyword.
Wherein, the preference value for initializing each keyword can be 1.When there is the identical keyword of preference value specific gravity, phase
Putting in order for keyword with preference value specific gravity can be with random alignment;It can also require to arrange according to certain, in the present embodiment
Without limitation.
Second news also may include the news that the first news of server push is concentrated, and also may include other news,
It the advantage is that, available bigger user's sample obtains the preference keyword of more accurate user;And reduce server
The operating pressure at end.
As optional, method further includes that server end executes following steps:
The preference value of each keyword of initialising subscriber;
According to the summation of the preference value of each keyword of user and the preference value of all keywords of user, obtain user's
The preference value specific gravity of each keyword;
The preference value specific gravity of each keyword of user is arranged from big to small, it is more default than before permutatation first to obtain preference value
The user preference keyword as the user of the keyword of value;
After the feedback information for obtaining user terminal;
Whether each news for judging that the first news is concentrated is browsed, and is increased by the preference value of the corresponding keyword of browsing news
1%, the preference value for the keyword not browsed reduces 1%;
The preference value specific gravity of each keyword of user is arranged from big to small, obtains preference value than the first preset value before permutatation
Keyword the user preference keyword as the user.
Wherein, the preference value for initializing each keyword can be 1.When there is the identical keyword of preference value specific gravity, phase
Putting in order for keyword with preference value specific gravity can be with random alignment;It can also require to arrange according to certain, in the present embodiment
Without limitation.
The big operating pressure for reducing user terminal can be beaten, the requirement of user terminal is reduced.
In step S3, it is based on user's news preference data and keyword coefficient of relationship, is obtained from wait push in news database
Take the news of the first default record as user preference news, comprising:
News to be pushed is obtained in news database from wait push, and extracts the keyword of respectively news to be pushed;
It is closed according to user's news that user preference keyword and the respectively keyword of news to be pushed calculate respectively news to be pushed
Join weight;
According to user's news associated weights of respectively news to be pushed, user's news associated weights are sorted from large to small, are obtained
Take the news to be pushed of the preceding first default record as user preference news.
It repeats to push in order to prevent, Xiang Tongyi user pushes when pushing news, has pushed and had pushed to same user
User's news associated weights of news to be pushed, are no longer ranked up.
In the present embodiment, the first default record can according to need setting, can be set to 5.If news to be pushed
User's news associated weights there are equal weights, can be randomly selected, such as the three of user's news associated weights ranking 4~6
A piece is identical, then can randomly choose 2 from this 3 is used as user's news associated weights the 4th and the 5th.
In order to make it easy to understand, herein to " news to be pushed for obtaining preceding first default record is new as user preference
It hears." do an explanation, when the first default record be 5 when, if user's news associated weights arrange from big to small respectively 0.100,
0.099,0.099,0.098,0.098,0.098,0.097,0.096,……;User's news associated weights be 0.100,0.099,
0.099 news to be pushed is respectively that the news and weight to be pushed of user's news associated weights the 1st, the 2nd, the 3rd are 0.98
Three wait push to be pushed news of random two news to be pushed as weight the 4th and the 5th in news, user's news is closed
The news to be pushed for joining weight the 1st, the 2nd, the 3rd, the 4th, the 5th is the news to be pushed of preceding 5 (the first default records), that is, uses
Family preference news.
Optionally, according to user preference keyword and respectively, the keyword of news to be pushed calculates the user of respectively news to be pushed
News associated weights include:
Obtain each keyword of news to be pushed and the coefficient of relationship of user preference keyword;
Each keyword for news to be pushed of summing and the coefficient of relationship of user preference keyword, obtain the use of news to be pushed
Family news associated weights.
Step S6: the feedback information based on user terminal, the keyword coefficient of relationship adjusted between keyword include:
Judge each news of the first news concentration whether clear by ustomer premises access equipment according to the feedback information of user terminal
It lookes at,
If the news that the first news is concentrated is browsed by ustomer premises access equipment, and the keyword by browsing news is opposite
Increase by 0.1% in the relation value of user preference keyword, and whether judge the relation value more than n, if relation value is more than n, incites somebody to action
The relation value is set to 1.
If the news that the first news is concentrated is not browsed by ustomer premises access equipment, by this not by the keyword phase of browsing news
Reduction 0.1% for the relation value of user preference keyword.
In present embodiment, by formula 1 it is recognised that coefficient of relationship YijN in/n is news overall number of keywords amount, if
There is new keyword, then n can increase, for example 1 new keyword occur, then n increases by 1.
Optionally, user's news preference data includes the preference value specific gravity of user preference keyword;
Feedback information based on user terminal, the keyword coefficient of relationship adjusted between keyword include:
Judge each news of the first news concentration whether clear by ustomer premises access equipment according to the feedback information of user terminal
It lookes at,
If the news that the first news is concentrated is browsed by ustomer premises access equipment, and the keyword by browsing news is opposite
Increase third preset value in the relation value of user preference keyword, and whether judges the relation value more than n, if relation value is more than n,
The relation value is then set to 1, the third preset value is the 0.1% of the preference value specific gravity of user preference keyword;
If the news that the first news is concentrated is not browsed by ustomer premises access equipment, by this not by the keyword phase of browsing news
For the 4th preset value of reduction of the relation value of user preference keyword, the 4th preset value is the inclined of user preference keyword
It is worth the 0.1% of specific gravity well.
When adjusting the keyword coefficient of relationship between keyword, the radix and preference value of adjustment compare re-association;It can be effective
Improve the reasonability of the keyword coefficient of relationship between adjustment keyword;Adjustment process is closer to reality.
Preferably, obtaining news to be pushed and its keyword includes: to establish referring to dictionary;News to be pushed is obtained, is mentioned
Take wait push the character in news, and by wait push the character in news with referring to being compared referring to keyword in dictionary,
If comparing successfully, will compare successfully referring to keyword as the preparation keyword of news to be pushed;By news to be pushed
Preparation keyword is compared with the character wait push in news, judges the number that prepared keyword occurs in news to be pushed
Fi, the number G that occurs in the title of news to be pushediAnd preparation keyword is in time occurred wait push each paragraph in news
Number Eij;Calculate the preparation keyword weight D that each prepared keyword is calculated by formula 2i;
In formula 2, DiIndicate the preparation keyword weight of preparation keyword i, FiIndicate preparation keyword i new wait push
Snuff existing number, GiIndicate the number G that preparation keyword i occurs in the title of news to be pushedi, EijIndicate that preparation is crucial
Word i indicates total paragraph number of news to be pushed in the number wait push the appearance of the jth section in news, n;
By the preparation keyword weight D of each prepared keywordiSort from large to small, to come preceding first preset quantity
Preparation keyword is used as news keyword to be pushed.First preset quantity can be arranged according to the actual situation, such as be set as 3,5
Deng.
It is known that, it include that can be preset referring to keyword referring to keyword referring to dictionary.
Present embodiment can extract the keyword in news, in case subsequent step uses;By referring to dictionary judge to
Pushing news, there are which possible keywords, i.e., prepared keyword, time occurred based on preparation keyword in news to be pushed
Number Fi, the number G that occurs in the title of news to be pushedi, and preparation keyword is wait push each paragraph appearance in news
Number Eij, the weight of each prepared keyword is calculated by formula 1, and the key of news to be pushed is obtained according to the sequence of weight
Word.It can also effectively identify the keyword of news to be pushed.
Due to the particularity of news, new keyword is easy to appear in news, and because Internet news is popularized, network
There are the news of more not set keyword in news, and there is also the characteristics of the more news provided with keyword;
The particularity of particularity and Internet news for news itself is obtained as the preferred of the disclosure wait push
News and its keyword include:
News to be pushed is obtained, is judged wait push in news whether keyword has been arranged;
If obtaining key of the keyword being arranged as news to be pushed wait push the keyword being arranged in news
Word, and be added to using the keyword being arranged as referring to keyword referring to dictionary;
If being extracted wait push the character in news, and will be wait push in news wait push not set keyword in news
Character with referring to being compared referring to keyword in dictionary, if comparing successfully, will compare successfully referring to keyword make
For the preparation keyword of news to be pushed;The preparation keyword of news to be pushed is compared with wait push the character in news
It is right, judge the number F that prepared keyword occurs in news to be pushedi, the number G that occurs in the title of news to be pushedi, with
And preparation keyword is wait push the number E that each paragraph occurs in newsij;It calculates and each prepared keyword is calculated by formula 2
Preparation keyword weight Di;
In formula 2, DiIndicate the preparation keyword weight of preparation keyword i, FiIndicate preparation keyword i new wait push
Snuff existing number, GiIndicate the number G that preparation keyword i occurs in the title of news to be pushedi, EijIndicate that preparation is crucial
Word i indicates total paragraph number of news to be pushed in the number wait push the appearance of the jth section in news, n;
By the preparation keyword weight D of each prepared keywordiSort from large to small, to come preceding first preset quantity
Preparation keyword is used as news keyword to be pushed.
In the present embodiment, for the characteristic in Internet news, according to the keyword of news to be pushed being arranged, automatically more
New optimization is referring to dictionary;When pushing news not set keyword, according to the reference dictionary optimized, automatic identification waits pushing
The keyword of news.Substantially increase the acquisition efficiency and accuracy rate of news keyword to be pushed.
In the description of this specification, reference term " one embodiment/mode ", " some embodiment/modes ", " show
The description of example ", " specific example " or " some examples " etc. mean to combine the specific features of the embodiment/mode or example description,
Structure, material or feature are contained at least one embodiment/mode or example of the application.In the present specification, to upper
The schematic representation for stating term is necessarily directed to identical embodiment/mode or example.Moreover, the specific features of description,
Structure, material or feature can be combined in any suitable manner in any one or more embodiment/modes or example.In addition,
Without conflicting with each other, those skilled in the art can by different embodiment/modes described in this specification or
Example and different embodiment/modes or exemplary feature are combined.
In addition, term " first ", " second " are used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance
Or implicitly indicate the quantity of indicated technical characteristic.Define " first " as a result, the feature of " second " can be expressed or
Implicitly include at least one this feature.In the description of the present application, the meaning of " plurality " is at least two, such as two, three
It is a etc., unless otherwise specifically defined.
It will be understood by those of skill in the art that above embodiment is used for the purpose of clearly demonstrating the disclosure, and simultaneously
Non- be defined to the scope of the present disclosure.For those skilled in the art, may be used also on the basis of disclosed above
To make other variations or modification, and these variations or modification are still in the scope of the present disclosure.
Claims (10)
1. a kind of dissemination of news method based on big data processing, which is characterized in that be included in server end and execute following steps:
News to be pushed and its keyword are obtained, news database to be pushed is formed according to news to be pushed and its keyword;
Wait news push trigger signal;
If receiving news push trigger signal, keyword coefficient of relationship between keyword and user to be pushed are obtained
User's news preference data, user's news preference data include User ID and user preference keyword;
Based on user's news preference data and keyword coefficient of relationship, the first default record is obtained in news database from wait push
News to be pushed as user preference news;
The news to be pushed for obtaining the second default record at random from news library to be pushed tests news as user;
The first news collection is pushed to user terminal, and obtains the feedback information of user terminal, wherein the first news collection includes
User preference news and user test news, and the feedback information of user terminal includes whether each news of the first news concentration is clear
The information look at;
Feedback information based on user terminal adjusts the keyword coefficient of relationship between keyword.
2. a kind of dissemination of news method based on big data processing as described in claim 1, which is characterized in that method further includes
Following steps are executed in server end:
By the keyword coefficient of relationship N between keywordijIt is initialized as follows,
In above formula, NijIndicate coefficient of relationship of i-th of keyword relative to j-th of keyword;When i=j, i-th of keyword phase
Coefficient of relationship for j-th of keyword is 1;When i ≠ j, i-th of keyword is relative to the coefficient of relationship of j-th of keywordWherein, YijIndicate relation value of i-th of keyword relative to j-th of keyword, n is keyword total quantity, YijInitial value
It is 1.
3. a kind of dissemination of news method based on big data processing as claimed in claim 2, which is characterized in that method further includes
Following steps are executed in user terminal:
Initialize the preference value of each keyword;
According to the summation of the preference value of each keyword and the preference value of all keywords, the preference value specific gravity of each keyword is obtained;
The preference value specific gravity of each keyword is arranged from big to small, obtains keyword of the preference value than the first preset value before permutatation
The user preference keyword as the user;
Receive the second news of simultaneously display server end push;
Judge whether the second news is browsed;
If the second news is browsed, the preference value which corresponds to keyword increases by 1%, otherwise, second news pair
The preference value of keyword is answered to reduce 1%;
The preference value specific gravity of each keyword is arranged from big to small, obtains keyword of the preference value than the first preset value before permutatation
The user preference keyword as the user.
4. a kind of dissemination of news method based on big data processing as claimed in claim 2, which is characterized in that method further includes
Following steps are executed in server end:
The preference value of each keyword of initialising subscriber;
According to the summation of the preference value of each keyword of user and the preference value of all keywords of user, each pass of user is obtained
The preference value specific gravity of keyword;
The preference value specific gravity of each keyword of user is arranged from big to small, obtains preference value than the first preset value before permutatation
The user preference keyword as the user of keyword;
After the feedback information for obtaining user terminal;
Whether each news for judging that the first news is concentrated is browsed, and increases by 1% by the preference value of the corresponding keyword of browsing news,
The preference value for the keyword not browsed reduces 1%;
The preference value specific gravity of each keyword of user is arranged from big to small, obtains pass of the preference value than the first preset value before permutatation
The user preference keyword as the user of keyword.
5. a kind of dissemination of news method based on big data processing as described in claim 3 or 4, which is characterized in that based on use
Family news preference data and keyword coefficient of relationship, from wait push the news conduct for obtaining the first default record in news database
User preference news, comprising:
News to be pushed is obtained in news database from wait push, and extracts the keyword of respectively news to be pushed;
Power is associated with user's news that the keyword of respectively news to be pushed calculates respectively news to be pushed according to user preference keyword
Weight;
According to user's news associated weights of respectively news to be pushed, user's news associated weights are sorted from large to small, before acquisition
The news to be pushed of first default record is as user preference news.
6. a kind of dissemination of news method based on big data processing as claimed in claim 5, which is characterized in that inclined according to user
User's news associated weights that the keyword of good keyword and respectively news to be pushed calculates respectively news to be pushed include:
Obtain each keyword of news to be pushed and the coefficient of relationship of user preference keyword;
Each keyword for news to be pushed of summing and the coefficient of relationship of user preference keyword, the user for obtaining news to be pushed are new
Hear associated weights.
7. a kind of dissemination of news method based on big data processing as claimed in claim 5, which is characterized in that whole based on user
The feedback information at end, the keyword coefficient of relationship adjusted between keyword include:
Judge whether each news that the first news is concentrated is browsed by ustomer premises access equipment according to the feedback information of user terminal,
If the first news concentrate news browsed by ustomer premises access equipment, by the keyword by browsing news relative to
The relation value of family preference keyword increases by 0.1%, and judges the relation value whether more than n, if relation value is more than n, by the pass
Set occurrence is set to 1;
If the first news concentrate news do not browsed by ustomer premises access equipment, by this not by the keyword of browsing news relative to
The reduction 0.1% of the relation value of user preference keyword.
8. a kind of dissemination of news method based on big data processing as claimed in claim 5, which is characterized in that the user is new
Hear the preference value specific gravity that preference data includes user preference keyword;
Feedback information based on user terminal, the keyword coefficient of relationship adjusted between keyword include:
Judge whether each news that the first news is concentrated is browsed by ustomer premises access equipment according to the feedback information of user terminal,
If the first news concentrate news browsed by ustomer premises access equipment, by the keyword by browsing news relative to
The relation value of family preference keyword increases third preset value, and whether judges the relation value more than n, if relation value is more than n, incites somebody to action
The relation value is set to 1, and the third preset value is the 0.1% of the preference value specific gravity of user preference keyword;
If the first news concentrate news do not browsed by ustomer premises access equipment, by this not by the keyword of browsing news relative to
The 4th preset value of reduction of the relation value of user preference keyword, the 4th preset value are the preference value of user preference keyword
The 0.1% of specific gravity.
9. a kind of dissemination of news method based on big data processing as described in claim 1, which is characterized in that a first default piece
Several and the second default record equal 5.
10. a kind of dissemination of news method based on big data processing as described in claim 1, which is characterized in that first is default
Value is 5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910833901.1A CN110532478B (en) | 2019-09-04 | 2019-09-04 | News dissemination method based on big data processing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910833901.1A CN110532478B (en) | 2019-09-04 | 2019-09-04 | News dissemination method based on big data processing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110532478A true CN110532478A (en) | 2019-12-03 |
CN110532478B CN110532478B (en) | 2022-05-03 |
Family
ID=68667024
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910833901.1A Active CN110532478B (en) | 2019-09-04 | 2019-09-04 | News dissemination method based on big data processing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110532478B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111970327A (en) * | 2020-07-22 | 2020-11-20 | 南京审计大学 | News spreading method and system based on big data processing |
CN112565342A (en) * | 2020-11-13 | 2021-03-26 | 广州市百果园网络科技有限公司 | Message pushing method, device, equipment and storage medium based on weight calculation |
CN117009672A (en) * | 2023-10-08 | 2023-11-07 | 江西科技学院 | Activity recommendation method and system based on big data |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102637171A (en) * | 2011-02-10 | 2012-08-15 | 北京百度网讯科技有限公司 | Method and device for optimizing searching results |
KR20140027011A (en) * | 2012-08-24 | 2014-03-06 | 삼성전자주식회사 | Method and server for recommending friends, and terminal thereof |
US20190026361A1 (en) * | 2017-07-24 | 2019-01-24 | Mycelebs Co., Ltd. | Method and apparatus for providing information by using degree of association between reserved word and attribute language |
CN109299383A (en) * | 2018-11-02 | 2019-02-01 | 北京字节跳动网络技术有限公司 | Generate method, apparatus, electronic equipment and the storage medium for recommending word |
CN109727047A (en) * | 2017-10-30 | 2019-05-07 | 北京京东尚科信息技术有限公司 | A kind of method and apparatus, data recommendation method and the device of determining data correlation degree |
CN110113410A (en) * | 2019-04-30 | 2019-08-09 | 秒针信息技术有限公司 | A kind of management method, device, electronic equipment and the storage medium of information push |
-
2019
- 2019-09-04 CN CN201910833901.1A patent/CN110532478B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102637171A (en) * | 2011-02-10 | 2012-08-15 | 北京百度网讯科技有限公司 | Method and device for optimizing searching results |
KR20140027011A (en) * | 2012-08-24 | 2014-03-06 | 삼성전자주식회사 | Method and server for recommending friends, and terminal thereof |
US20190026361A1 (en) * | 2017-07-24 | 2019-01-24 | Mycelebs Co., Ltd. | Method and apparatus for providing information by using degree of association between reserved word and attribute language |
CN109727047A (en) * | 2017-10-30 | 2019-05-07 | 北京京东尚科信息技术有限公司 | A kind of method and apparatus, data recommendation method and the device of determining data correlation degree |
CN109299383A (en) * | 2018-11-02 | 2019-02-01 | 北京字节跳动网络技术有限公司 | Generate method, apparatus, electronic equipment and the storage medium for recommending word |
CN110113410A (en) * | 2019-04-30 | 2019-08-09 | 秒针信息技术有限公司 | A kind of management method, device, electronic equipment and the storage medium of information push |
Non-Patent Citations (1)
Title |
---|
李晓黎 等: "高校图书情报机构信息服务中应用推送技术的探讨", 《高校图书馆工作》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111970327A (en) * | 2020-07-22 | 2020-11-20 | 南京审计大学 | News spreading method and system based on big data processing |
CN112565342A (en) * | 2020-11-13 | 2021-03-26 | 广州市百果园网络科技有限公司 | Message pushing method, device, equipment and storage medium based on weight calculation |
CN117009672A (en) * | 2023-10-08 | 2023-11-07 | 江西科技学院 | Activity recommendation method and system based on big data |
CN117009672B (en) * | 2023-10-08 | 2024-01-09 | 江西科技学院 | Activity recommendation method and system based on big data |
Also Published As
Publication number | Publication date |
---|---|
CN110532478B (en) | 2022-05-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110532478A (en) | A kind of dissemination of news method based on big data processing | |
KR101532054B1 (en) | Blending mobile search results | |
CN101609472B (en) | Keyword evaluation method and device based on platform for questions and answers | |
CN103488788B (en) | It is a kind of to apply the method and device recommended automatically | |
CN110413868B (en) | Information recommendation method, device, system and storage medium | |
WO2019076191A1 (en) | Keyword extraction method and device, and storage medium and electronic device | |
CN104008109A (en) | User interest based Web information push service system | |
CN103593425B (en) | Preference-based intelligent retrieval method and system | |
CN103186574B (en) | A kind of generation method and apparatus of Search Results | |
CN102163229B (en) | Method and equipment for generating abstracts of searching results | |
CN106570144A (en) | Method and apparatus for recommending information | |
CN106250526A (en) | A kind of text class based on content and user behavior recommends method and apparatus | |
CN101876981A (en) | Method and device for establishing knowledge base | |
US20110184930A1 (en) | Methods and Systems for Adjusting a Scoring Measure Based on Query Breadth | |
CN103729359A (en) | Method and system for recommending search terms | |
CN105138558A (en) | User access content-based real-time personalized information collection method | |
CN107885717B (en) | Keyword extraction method and device | |
CN109933708A (en) | Information retrieval method, device, storage medium and computer equipment | |
CN111159404A (en) | Text classification method and device | |
CN104750760A (en) | Application software recommending method and device | |
CN102364467A (en) | Network search method and system | |
CN103279504A (en) | Searching method and device based on ambiguity resolution | |
CN102999538B (en) | Personage's searching method and equipment | |
WO2010037314A1 (en) | A method for searching and the device and system thereof | |
CN111192170B (en) | Question pushing method, device, equipment and computer readable storage medium |
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 |