CN113609424A - Computing and early warning system and method for network public sentiment popularity - Google Patents
Computing and early warning system and method for network public sentiment popularity Download PDFInfo
- Publication number
- CN113609424A CN113609424A CN202110693652.8A CN202110693652A CN113609424A CN 113609424 A CN113609424 A CN 113609424A CN 202110693652 A CN202110693652 A CN 202110693652A CN 113609424 A CN113609424 A CN 113609424A
- Authority
- CN
- China
- Prior art keywords
- news
- heat
- popularity
- media
- report
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000004364 calculation method Methods 0.000 claims abstract description 50
- 230000000694 effects Effects 0.000 claims abstract description 27
- 230000002452 interceptive effect Effects 0.000 claims abstract description 26
- 230000003993 interaction Effects 0.000 claims description 18
- 238000010606 normalization Methods 0.000 claims description 7
- 230000006399 behavior Effects 0.000 claims description 3
- 230000001186 cumulative effect Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 2
- 238000012552 review Methods 0.000 claims description 2
- 238000004458 analytical method Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 235000019633 pungent taste Nutrition 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
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
- G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Abstract
The invention discloses a computing and early warning system and method aiming at network public sentiment popularity, and relates to the technical field of network security. The news media popularity module consists of a report popularity unit, a report coverage unit and a report authority unit and is used for reflecting the popularity of the network public sentiment; the social platform popularity module is composed of a posting popularity unit, an interactive popularity unit, an account activity unit and a publisher influence unit and is used for reflecting the popularity of the network. The method solves the problem that the information density is too low due to the fact that data acquired from the whole network in the traditional network public opinion popularity heat calculation method is too wide. The method solves the problem that the influence and authority degree of different media and different users are not considered to be different when the traditional network public opinion popularity heat index is calculated. The method solves the problems that when the popularity of the traditional network is calculated, the weight definition is mixed with artificial subjectivity and cannot be updated aiming at different topics.
Description
Technical Field
The invention relates to the technical field of network security, in particular to a computing and early warning system and method for network public opinion popularity heat.
Background
The popularity of the network is reflected by the media reports and the concerns and discussion degree of netizens caused by the events occurring in the real society on the network in a certain time period, and is one of the important indexes for network public opinion analysis and public opinion early warning.
At present, two methods are generally used for calculating the popularity of network public opinion:
the first is an analysis method based on data mining, firstly, massive data are collected from social media or other network platforms through an information collection technology, then a computer-easy-to-process form is obtained through technologies such as data preprocessing and the like, and finally, the obtained data are clustered through methods such as text similarity calculation and the like by utilizing a natural language processing technology, so that topic events with the highest attention degree and the largest influence are obtained, and the heat degree of topics is calculated;
the second public sentiment popularity calculating method based on the content is divided into methods based on media, based on users and combined analysis of users and media, and the public sentiment popularity analysis based on media angles calculates the weighted sum of the values by counting the number of reports, the report speed, the report days and the report frequency of a certain hot event on some media platforms within a period of time and then assigning different weights to the values, thereby calculating the popularity of the hot event; the public opinion popularity degree analysis based on the user angle firstly calculates the number of posts collected about a certain topic, the browsing number, the forwarding number, the comment number and the return number of each post, and then calculates the popularity degree of the certain topic by using the data.
The existing method for calculating the popularity of the network public opinion has the defects of several aspects:
(1) at present, no calculation method for the popularity of the network public sentiment exists, so that the calculation accuracy of the popularity of the network public sentiment is low, and the reference for public sentiment early warning is poor. The concrete embodiment is as follows: the media and social platform with influence in each region are different, for example, the zhongjiang network mainly provides important news information in Jiangsu province, and has important public opinion influence in Jiangsu province, but the existing public opinion popularity calculation method does not consider the point in analysis.
(2) In the existing public opinion popularity heat calculation method based on content, the influence of different media is not considered when calculating from the media perspective; when the online public opinion popularity heat degree is analyzed from the perspective of the user, the user with high activity degree and high influence is not considered, and the accuracy of online public opinion heat degree calculation and early warning can be reduced. The concrete embodiment is as follows: people's network will produce bigger influence to public opinion heat than local media, and active users will also produce bigger influence to public opinion, which is not considered in the existing network public opinion heat calculation method.
(3) The network public opinion popularity is dynamically changed, and when the weight of each influence factor is determined by the conventional network public opinion popularity calculation mode, the public opinion popularity analysis early warning and the timeliness of tracking hot topics with potential risks cannot be guaranteed by generally adopting an expert questionnaire form or a specific analysis form aiming at specific topics.
(4) The prior method for calculating the popularity of the network public sentiment generally adopts absolute values when calculating each index, which can cause that the popularity of different types of network public sentiment topics has large difference and cannot be applied to the setting of a threshold value in public sentiment early warning.
Disclosure of Invention
In order to overcome the problems in the related art, the disclosed embodiment of the invention provides a system and a method for calculating and early warning of network public sentiment popularity. The technical scheme is as follows:
this calculation and early warning system to network public opinion popularity degree includes:
the news media popularity module consists of a report popularity unit, a report coverage unit and a report authority unit and is used for reflecting the popularity of the network public opinion;
the social platform popularity module is composed of a posting popularity unit, an interactive popularity unit, an account activity unit and a publisher influence unit and is used for reflecting the popularity of the network.
In one embodiment, the report popularity unit counts the number of published reports of the news media, the publishing rate of the news media, the number of days the news reports last, and the interactive popularity data of the news reports in a period of time;
a report coverage unit for counting the participation decision data of the central media;
the report authority unit, the participation degree of the statistical important media determines the data.
In one embodiment, the posting popularity unit is used for counting the number of posts posted by a user in a period of time;
the interactive popularity unit is used for counting the influence degrees of the user on the forwarding, praise and comment behaviors of the posts;
the account activity unit is used for counting the participation degree of the network users which are relatively active locally;
and the publisher influence degree unit is used for counting the participation degree of key characters in the acquired information.
Another objective of the present invention is to provide a method for implementing the system for calculating and warning popularity of internet, which is applied to an information data processing terminal, the method for calculating and warning popularity of internet comprises the following steps:
step one, selecting a news media platform with certain influence and propagation degree and a social network platform widely used in the local;
secondly, carrying out post information statistics and user information statistics on data obtained from the network social platform, and then calculating the posting heat, the interaction heat, the account activity and the publisher influence degree of the network social platform by using the data;
step three, the statistical information of the user is mainly used for calculating account activity and figure influence degree;
step four, the network public opinion popularity heat of the social platform is equal to the weighted sum of the post heat, the interaction heat, the account activity and the publisher influence after normalization;
calculating the network public opinion popularity of the news media platform based on the news media statistical information and the statistical information reported by the news media platform;
and step six, calculating the network public sentiment popularity of the current day through the calculation of the network public sentiment popularity of the social platform and the network public sentiment popularity of the news media.
In one embodiment, in the first step, when determining the list of important news media platforms, the top 100 news media platforms screened out according to the service field range and the existing authority ranking are determined as important news media and are continuously updated; when a key user list of the social platform is determined, firstly bringing known large V, high-influence users and authoritative users in the service field into the key list, subsequently obtaining the high-influence users and bridge users developing in each event according to the events, and dynamically updating the high-influence users and bridge users into an important user name list library of the social platform; and establishing a social platform important user database by collecting posting data and personal data of an important user name list library.
In one embodiment, in the step two, the interaction popularity needs to be obtained by calculating a weighted sum of the approval popularity, the forwarding popularity and the review popularity, and a specific calculation flow is as follows:
(1) the posting heat degree is equal to the ratio of the posting total amount of the current day to the latest maximum posting number obtained by statistics, and the specific formula is as follows:
wherein post _ heat represents the hot degree of posting of the day, post _ numtodayRepresents the number of posts posted on that day, max (post _ num)recent) A ratio representing the maximum number of recent posts;
(2) the comment popularity, the forward popularity and the comment popularity are calculated by the post information obtained by statistics, and the specific calculation formula is as follows:
interact_heat＝x*(like_heat)+y*(forward_heat)+z*(comment_heat)
wherein, interactive _ heat represents the interactive heat of the current day, like _ heat represents the praise heat of the current day, forward _ heat represents the forwarding heat of the current day, comment _ heat represents the comment heat of the current day, and x, y and z are weights obtained by an entropy weight method;
wherein (Sigma)Each postforwar_dnum)todRepresents the forwarded sum of all posts of the day, max (Σ)Each postlike_num)recentA maximum value representing a praise cumulative sum of recently calculated posts; (Sigma)Each postforward_num)todayRepresents the forwarded sum of all posts of the day, max (Σ)Each postforward_num)recentA maximum value representing a forwarding accumulated sum of recently calculated posts; (Sigma)Each postcomme_ntn)todThe comment sum, max (Σ), representing all posts of the dayEach postcomment_num)recentRepresenting a maximum of the accumulated sum of comments for a recently computed post.
In one embodiment, in step three, the specific calculation formula of the account activity is as follows:
wherein, users __ heat represents the activity of the user on the day, (active _ users)todayIndicating participation in discussion on the same dayActive user number (active _ users)allRepresenting the number of all the users marked as active users in the past;
the specific calculation formula of the influence degree of the publisher is as follows:
wherein, users _ effect represents the influence degree of the publisher, (interactive _ users)todayRepresents the key figure participating in the discussion on the day, (active _ users)allIndicating the total number of highlighted characters targeted,
indicating the degree of participation, x, of the previously targeted key character on the same dayiRepresenting the weight, y, occupied by the ith key character participating in the current day's discussioniIndicating the number of posts made by the ith key character on the current day.
In one embodiment, in step four, the weight of the weighted sum is determined by entropy weighting, and the specific calculation formula is as follows:
wherein, (social _ heat)todayRepresents the network public opinion popularity of the current day of the social platform,
represents the normalized heat of posting,
represents the interactive heat degree after the normalization,
representing the normalized user liveness,
the normalized publisher influence is expressed, and a, b, c, and d are weights obtained by an entropy weight method.
In one embodiment, in step five, the internet public opinion popularity of the news media platform is calculated as follows:
(1) calculating a report authority degree based on the news media statistical information, wherein the report authority degree is obtained by calculating important media participation degree and central-level media participation degree; the specific calculation method is as follows:
wherein, the participants _ coverage represents the participation degree of the important media, (the participants _ par)todayRepresenting the number of important media participating in news reports on the same day, representing the total number of the important news media, and determining the important news media before calculation according to the ranking of a news media website on an Alexa website, the total number of the news reports sent by the news media website, the attention number, the fan number and the reading number of the news media website;
wherein CC _ websites _ par represents the middle level media engagement, (CC _ websites _ par)todayCentral level media number (CC _ websites) representing the day's participation in news storiescountRepresenting the total number of central level media counted, not all central level media will be considered;
wherein news _ authority represents the report authority, (CC _ websites _ news)iThe number of news reports sent by the ith central level news media participating in the news reports on the same day is represented; (sites _ news)jThe story from the jth news platform marked as important media representing the current day's participation in news storiesNumber, miWeight corresponding to ith Central News media, njThe weight corresponding to the jth important media is determined according to the influence and the ranking on Alexa of each central level media and important media before calculation, and m and n are the number of the central level media and the number of the important media participating in the news report of the current day respectively;
(2) the report popularity is calculated based on the statistical information of news reports released by the news media and is equal to the weighted sum of the report popularity and the report interaction popularity; the report popularity is calculated by the report quantity and the report rate, and the report interaction popularity is calculated by the forwarding number, the praise number and the comment number of the news reports in a weighted manner in the following specific way:
wherein news _ heat represents news report heat, news _ numtodayIndicates the number of news reports of the day, max (news _ num)recent) Indicates the maximum number of recent news reports, max (news _ avg _ hour)todayIndicates the maximum number of news reports per hour of the day, i.e., the news reporting rate, max (news _ avg _ hour)recentRepresenting the most recent news reporting rate;
news_inter_heat＝x*(news_like_heat)+y*(news_forw_heat)+z*(news_com_heat)
wherein, news _ inter _ heat represents the interactive heat of the news report, news _ like _ heat represents the like heat of the news report, news _ forew _ heat represents the forwarding heat of the news report, news _ com _ heat represents the comment heat of the news report, and x, y and z are weights obtained by an entropy weight method;
(3) the network public opinion popularity heat of the news media platform is obtained by normalizing the report authority and the report heat and then calculating a weighted sum, the weights of the report authority and the report heat are determined by an entropy weight method, and the specific calculation formula is as follows:
wherein, (news _ web _ heat)todayRepresents the internet public opinion popularity of the news media of the current day,
showing the normalized report authority,
Indicates the normalized reported heat,
And expressing the normalized report interaction heat.
In one embodiment, in step six, the calculation formula of the internet public opinion popularity of the current day is as follows:
xx_heattoday＝(news_web_heat)today+(social_heat)today
wherein xx heattodayAnd the network public opinion popularity of the current day is represented.
The technical scheme provided by the embodiment of the invention has the following beneficial effects:
1. the method solves the problem that the information density is too low due to the fact that data acquired from the whole network in the traditional network public opinion popularity heat calculation method is too wide.
2. The problem that influence and authority of different media and different users are different when traditional online public opinion popularity indexes are calculated is solved, and online public opinion popularity brought by the authority media and opinion leaders accounts for more than 80% of the overall public opinion popularity.
3. The method solves the problems that when the popularity of the traditional network is calculated, the weight definition is mixed with artificial subjectivity and cannot be updated aiming at different topics. The method for calculating the popularity of the network public sentiment verifies and calculates the popularity of the public sentiment from 5 month 20 day to 8 month 1 day in a certain area in 2020 by using a traditional method for calculating the popularity of the network public sentiment and the method for calculating the popularity of the network public sentiment disclosed by the embodiment, and the time coincidence degree of the method for calculating the popularity of the network public sentiment disclosed by the embodiment and the large affair is higher than the time coincidence degree of the traditional method by more than 20 percent.
4. Public sentiment popularity is measured by utilizing the relative value, and the problem that the popularity early warning threshold value is difficult to determine in the traditional network public sentiment early warning prediction method is solved. The method comprises the steps of calculating the popularity of the network public sentiment in real time by using data acquired from news media and a social platform, reflecting the change of the popularity of the network public sentiment in time, carrying out early warning and tracking on some hot public sentiment topics with potential risks, and providing valuable references for relevant departments.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure.
Fig. 1 is an architecture diagram of an internet public opinion popularity index system provided by the present invention.
Fig. 2 is a flow chart of calculating popularity of network according to the present invention.
Fig. 3 is a flow chart of internet public opinion popularity heat analysis and early warning provided by the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, but rather should be construed as broadly as the present invention is capable of modification in various respects, all without departing from the spirit and scope of the present invention.
It will be understood that when an element is referred to as being "secured to" another element, it can be directly on the other element or intervening elements may also be present. When an element is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements may also be present. As used herein, the terms "vertical," "horizontal," "left," "right," and the like are for purposes of illustration only and are not intended to represent the only embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
The innovation points of the invention mainly comprise the following aspects:
(1) in the aspect of data acquisition, the invention aims at the network public opinion data acquisition scheme, and builds a list library of key characters of a news media platform and a social network platform. For a news media platform, acquiring data of 100 news media platforms which have great influence in the area in real time every day; for the social network platform, after preliminary analysis, data of users with high liveness and large influence in a relatively long period of time are collected, and the collected data can represent local network public opinion conditions better and contain high information density.
(2) In the aspect of constructing an index system, the index system which can be applied to the calculation of the popularity of the network is constructed. The weight updating rule based on the entropy weight method is innovatively provided when the posting text and the news data are calculated, and the subjective error caused by artificially defining the weight is avoided.
(3) When an important news media and an important user name list library of a social network platform are constructed, a dynamic updating method is adopted. When a list of key news media platforms is determined, the top 100 news media platforms screened out according to the service field range and the existing authority ranking are determined as key news media and are continuously updated; when the key user list of the social platform is determined, known large V users, high-influence users and authoritative users in the service field are brought into the key list, the high-influence users and bridge users developing in each event are obtained according to the events, and the high-influence users and the bridge users are dynamically updated into the key user list library of the social platform. And establishing a social platform important user database by collecting posting data and personal data of an important user name list library.
The network public opinion popularity heat index system architecture mainly comprises the following aspects:
1) news media
The news media is an important channel for network information propagation, is an important index for reflecting the popularity of the network, and mainly comprises the report popularity, the report coverage and the report authority. Wherein the report popularity is composed of the number of published reports of the news media, the speed of publishing the reports of the news media, the number of days the news reports last, and the interactive popularity of the news reports over a period of time; the reporting authority is mainly determined by the participation of the central media; coverage is determined by the participation level of the important media.
2) Social platform
The social platform is the most direct sounding place of a character on the network, and is another important index reflecting the popularity of the network. The method mainly comprises the following steps of posting heat degree, interaction heat degree, account activity degree and publisher influence degree. Wherein the posting hotness is determined by the number of posts posted by the user within a period of time; the interactive heat is influenced by the behaviors of forwarding, praise, comment and the like of the user on the posts; account activity is mainly reflected in the participation of local relatively active network users; the publisher influence degree is formed by the participation degree of key people in the collected information.
The network public sentiment popularity is dynamically changed, so the method is calculated based on the related data of the network public sentiment popularity on each mainstream social platform and specific news media in a period of time. The main flow of the network public opinion popularity heat calculation and early warning is as follows:
1. selecting a news media platform with certain influence and spreading degree and a social network platform which is widely used locally, wherein the specific method is that when a list of key news media platforms is determined, the top 100 news media platforms are screened out according to the service field range and the existing authority ranking and determined as key news media and are continuously updated; when the key user list of the social platform is determined, known large V users, high-influence users and authoritative users in the service field are brought into the key list, the high-influence users and bridge users developing in each event are obtained according to the events, and the high-influence users and the bridge users are dynamically updated into the key user list library of the social platform. And establishing a social platform important user database by collecting posting data and personal data of an important user name list library.
2. Performing post information statistics and user information statistics on data obtained from a network social platform, wherein the post information statistics mainly comprises the statistics of the number of posts, the forwarding number of the posts, the number of praise and the number of comments; the user information statistics mainly comprises the steps of counting the daily average posting number of the user, the fan number of the user and the concerned number of the user, and then calculating the posting heat, the interaction heat, the account activity and the publisher influence degree of the network social platform by using the data. The interactive popularity needs to be obtained by calculating the weighted sum of the endorsement popularity, the forwarding popularity and the comment popularity, and the specific calculation flow is as follows:
(1) the posting heat degree is equal to the ratio of the posting total amount of the current day to the latest maximum posting number obtained by statistics, and the specific formula is as follows:
wherein post _ heat represents the hot degree of posting of the day, post _ numtodayRepresents the number of posts posted on that day, max (post _ num)recent) Representing the ratio of the maximum number of recent posts.
(2) The interactive popularity is equal to the weighted sum of the praise popularity, the forward popularity and the comment popularity, and the weight is obtained by an entropy weight method. Wherein, the popularity of points, the popularity of forwarding and the popularity of comments are calculated by the post information obtained by statistics. The specific calculation formula is as follows:
interact_heat＝x*(like_heat)+y*(forward_heat)+z*(comment_heat)
the interactive _ heat represents the interactive heat of the current day, the like _ heat represents the praise heat of the current day, the forward _ heat represents the forwarding heat of the current day, the comment _ heat represents the comment heat of the current day, and x, y and z are weights obtained by an entropy weight method.
Wherein (Sigma)Each postforwar_dnum)todRepresents the forwarded sum of all posts of the day, max (Σ)Each postlike_num)recentA maximum value representing a praise cumulative sum of recently calculated posts; (Sigma)Each postforward_num)todayRepresents the forwarded sum of all posts of the day, max (Σ)Each postforward_num)recentA maximum value representing a forwarding accumulated sum of recently calculated posts; (Sigma)Each postcomme_ntn)todThe comment sum, max (Σ), representing all posts of the dayEach postcomment_num)recentRepresenting a maximum of the accumulated sum of comments for a recently computed post.
3. The statistical information of the user is mainly used for calculating account activity and character influence degree.
(1) The current day's account activity refers to the ratio of users participating in the discussion on the current day who have been determined to be active accounts to all active users. The past active user group is determined according to posting and comment frequency of the user within a period of time, when the arithmetic sum of the posting and comment frequency exceeds a certain threshold value, the active user is marked, and users who post and comment frequency arithmetic and rank 10000 before the key user name list library are taken as active users. The specific calculation formula of the account activity is as follows:
wherein, users __ heat represents the activity of the user on the day, (active _ users)todayRepresents the active user number participating in the discussion on the same day (active _ users)allIndicating the number of all users that were previously rated as active users.
(2) The publisher influence degree is calculated by the key character participation degree and the posting number of the key character, wherein the key character participation degree refers to the proportion of users marked as the key character participating in discussion in the same day, the key character is determined according to the number of fans, the attention number, the reading number of posted posts and the forwarding number in statistical information, and specifically, the number of fans, the attention number, the reading number of posted posts, the forwarding number in the key name list library are weighted and ranked as the former 10000 users. The specific calculation formula of the influence degree of the publisher is as follows:
wherein, users _ effect represents the influence degree of the publisher, (interactive _ users)todayRepresents the key figure participating in the discussion on the day, (active _ users)allIndicating the total number of highlighted characters targeted,
indicating the degree of participation, x, of the previously targeted key character on the same dayiRepresenting the weight, y, occupied by the ith key character participating in the current day's discussioniIndicating the number of posts made by the ith key character on the current day.
4. The network public opinion popularity heat of the social platform is equal to the weighted sum of post heat, interaction heat, account activity and publisher influence after normalization, wherein the weight is determined by an entropy weight method, and a specific calculation formula is as follows:
wherein, (social _ heat)todayRepresents the network public opinion popularity of the current day of the social platform,
represents the normalized heat of posting,
represents the interactive heat degree after the normalization,
representing the normalized user liveness,
the normalized publisher influence is expressed, and a, b, c, and d are weights obtained by an entropy weight method.
5. The network public opinion popularity of the news media platform is calculated based on the news media statistical information and the statistical information of the news channel sent by the news media platform. The news media statistical information mainly comprises whether a news media website is a central-level media, the ranking of the news media website on an Alexa website, the total quantity of news reports sent by the news media website, the number of concerns, the number of fans and the number of readings of the news media website, and the statistical information of the reports sent by the news media website mainly comprises the number of recent news reports, the number of reading of each news report, the number of forwarding, the number of praise and the number of comments. The specific calculation method is as follows:
(1) the story authority is calculated based on the news media statistics. The method is specifically calculated by the important media participation and the central-level media participation. The specific calculation method is as follows:
wherein, the participants _ coverage represents the participation degree of the important media, (the participants _ par)todayImportant media representing the participation in news stories on the same dayAnd the important _ news represents the total number of important news media, and the important news media is determined before calculation according to the ranking of the news media website on the Alexa website, the total amount of news reports sent by the news media website, the attention number, the fan number and the reading number of the news media website.
Wherein CC _ websites _ par represents the middle level media engagement, (CC _ websites _ par)todayCentral level media number (CC _ websites) representing the day's participation in news storiescountRepresenting the total number of center level media counted, not all center level media will be considered.
Wherein news _ authority represents the report authority, (CC _ websites _ news)iIndicating the number of news reports sent by the ith central level news media participating in the news reports on the same day. (sites _ news)jNumber of stories, m, from the jth news platform targeted as important media that represents the current day's participation in news storiesiWeight corresponding to ith Central News media, njThe weight corresponding to the jth important media is determined according to the influence and the ranking on Alexa of each central level media and important media before calculation, and m and n are the number of central level media and the number of important media participating in the news report of the current day respectively.
(2) The report popularity is calculated based on the statistical information of the news reports released by the news media and is equal to the weighted sum of the report popularity and the report interaction popularity. The report popularity is obtained by calculating the report quantity and the report rate, and the report interaction popularity is obtained by calculating the weighted sum of the forwarding number, the praise number and the comment number of the news reports, and the specific calculation mode is as follows:
wherein news _ heat represents news report heat, news _ numtodayIndicates the number of news reports of the day, max (news _ num)recent) Indicates the maximum number of recent news reports, max (news _ avg _ hour)todayIndicates the maximum number of news reports per hour of the day, i.e., the news reporting rate, max (news _ avg _ hour)recentIndicating the maximum recent news reporting rate.
news_inter_heat＝x*(news_like_heat)+y*(news_forw_heat)+z*(news_com_heat)
Wherein news _ inter _ heat represents the interactive heat of the news report, news _ like _ heat represents the like heat of the news report, news _ forew _ heat represents the forwarding heat of the news report, news _ com _ heat represents the comment heat of the news report, and x, y and z are weights obtained by an entropy weight method.
(3) The network public opinion popularity heat of the news media platform is obtained by weighting and calculating after normalization of the report authority and the report heat, and the weights of the report authority and the report heat are determined by an entropy weight method. The specific calculation formula is as follows:
wherein, (news _ web _ heat)todayRepresents the internet public opinion popularity of the news media of the current day,
showing the normalized report authority,
Indicates the normalized reported heat,
And expressing the normalized report interaction heat.
6. The network public sentiment popularity of the current day is finally equal to the arithmetic sum of the network public sentiment popularity of the social platform and the network public sentiment popularity of the news media. The specific calculation formula is as follows:
xx_heattoday＝(news_web_heat)today+(social_heat)today
wherein xx heattodayAnd the network public opinion popularity of the current day is represented.
Compared with the traditional mode of directly constructing an index system and determining the weight according to post forwarding, praise and comment, the method combines business expert evaluation and actual occurrence of events, and improves the contact ratio with the actual occurrence of major events by 12% after multiple practical tests.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure should be limited only by the attached claims.
Claims (10)
1. A calculation and early warning system aiming at network public sentiment popularity is characterized in that the calculation and early warning system aiming at the network public sentiment popularity comprises:
the news media popularity module consists of a report popularity unit, a report coverage unit and a report authority unit and is used for reflecting the popularity of the network public opinion;
the social platform popularity module is composed of a posting popularity unit, an interactive popularity unit, an account activity unit and a publisher influence unit and is used for reflecting the popularity of the network.
2. The system of claim 1, wherein the report popularity unit is configured to count a number of published reports of news media, a rate of the published reports of the news media, a number of days that the news reports last, and interactive popularity data of the news reports over a period of time;
a report coverage unit for counting the participation decision data of the central media;
the report authority unit, the participation degree of the statistical important media determines the data.
3. The system of claim 1, wherein the posting popularity unit is configured to count the number of posts posted by the user over a period of time;
the interactive popularity unit is used for counting the influence degrees of the user on the forwarding, praise and comment behaviors of the posts;
the account activity unit is used for counting the participation degree of the network users which are relatively active locally;
and the publisher influence degree unit is used for counting the participation degree of key characters in the acquired information.
4. A method for implementing the system for calculating and warning popularity of internet according to any one of claims 1 to 3, wherein the method is applied to an information data processing terminal, and the method for calculating and warning popularity of internet comprises the following steps:
step one, selecting a news media platform with certain influence and propagation degree and a social network platform widely used in the local;
secondly, carrying out post information statistics and user information statistics on data obtained from the network social platform, and then calculating the posting heat, the interaction heat, the account activity and the publisher influence degree of the network social platform by using the data;
step three, calculating account activity and figure influence degree by using the statistical information of the user;
step four, the network public opinion popularity heat of the social platform is equal to the weighted sum of the post heat, the interaction heat, the account activity and the publisher influence after normalization;
calculating the network public opinion popularity of the news media platform based on the news media statistical information and the statistical information reported by the news media platform;
and step six, calculating the network public sentiment popularity of the current day through the calculation of the network public sentiment popularity of the social platform and the network public sentiment popularity of the news media.
5. The method for calculating and warning the popularity of internet according to claim 4, wherein in the first step, when determining the list of important news media platforms, the top 100 news media platforms screened out according to the business domain range and the existing authority ranking are determined as important news media and are continuously updated; when a key user list of the social platform is determined, firstly bringing known large V, high-influence users and authoritative users in the service field into the key list, subsequently obtaining the high-influence users and bridge users developing in each event according to the events, and dynamically updating the high-influence users and bridge users into an important user name list library of the social platform; and establishing a social platform important user database by collecting posting data and personal data of an important user name list library.
6. The method for calculating and warning network public opinion popularity heat according to claim 4, wherein in the second step, the interaction heat is obtained by calculating a weighted sum of like heat, forward heat, and review heat, and the specific calculation flow is as follows:
(1) the posting heat degree is equal to the ratio of the posting total amount of the current day to the latest maximum posting number obtained by statistics, and the specific formula is as follows:
wherein post _ heat represents the hot degree of posting of the day, post _ numtodayRepresents the number of posts posted on that day, max (post _ num)recent) Watch (A)Showing the maximum number of posts in the recent period;
(2) the comment popularity, the forward popularity and the comment popularity are calculated by the post information obtained by statistics, and the specific calculation formula is as follows:
interact_heat＝x*(like_heat)+y*(forward_heat)+z*(comment_heat)
wherein, interactive _ heat represents the interactive heat of the current day, like _ heat represents the praise heat of the current day, forward _ heat represents the forwarding heat of the current day, comment _ heat represents the comment heat of the current day, and x, y and z are weights obtained by an entropy weight method;
wherein (Sigma)Each postforward_num)todayRepresents the forwarded sum of all posts of the day, max (Σ)Each postlike_num)recentA maximum value representing a praise cumulative sum of recently calculated posts; (Sigma)Each postforward_num)todayRepresents the forwarded sum of all posts of the day, max (Σ)Each postforward_num)recentA maximum value representing a forwarding accumulated sum of recently calculated posts; (Sigma)Each postcomment_num)todayThe comment sum, max (Σ), representing all posts of the dayEach postcomment_num)recentRepresenting a maximum of the accumulated sum of comments for a recently computed post.
7. The method for calculating and warning of internet public opinion popularity heat according to claim 4, wherein in step three, the specific calculation formula of the account activity is as follows:
wherein, users __ heat represents the activity of the user on the day, (active _ users)todayRepresents the active user number participating in the discussion on the same day (active _ users)allRepresenting the number of all the users marked as active users in the past;
the specific calculation formula of the influence degree of the publisher is as follows:
wherein, users _ effect represents the influence degree of the publisher, (interactive _ users)todayRepresents the key figure participating in the discussion on the day, (active _ users)allIndicating the total number of highlighted characters targeted,
indicating the degree of participation, x, of the previously targeted key character on the same dayiRepresenting the weight, y, occupied by the ith key character participating in the current day's discussioniIndicating the number of posts made by the ith key character on the current day.
8. The method for calculating and warning of network public opinion popularity heat according to claim 4, wherein in step four, the weight is determined by an entropy weight method, and the specific calculation formula is as follows:
wherein, (social _ heat)todayRepresents the network public opinion popularity of the current day of the social platform,
represents the normalized heat of posting,
represents the interactive heat degree after the normalization,
representing the normalized user liveness,
the normalized publisher influence is expressed, and a, b, c, and d are weights obtained by an entropy weight method.
9. The method for calculating and warning network public opinion popularity heat according to claim 4, wherein in the fifth step, the network public opinion heat of the news media platform is calculated as follows:
(1) calculating a report authority degree based on the news media statistical information, wherein the report authority degree is obtained by calculating important media participation degree and central-level media participation degree; the specific calculation method is as follows:
wherein, the participants _ coverage represents the participation degree of the important media, (the participants _ par)todayRepresenting the number of important media participating in news reports on the same day, representing the total number of the important news media, and determining the important news media before calculation according to the ranking of a news media website on an Alexa website, the total number of the news reports sent by the news media website, the attention number, the fan number and the reading number of the news media website;
wherein CC _ websites _ par represents the middle level media engagement, (CC _ websites _ par)todayCentral level media number (CC _ websites) representing the day's participation in news storiescountRepresenting the total number of central level media counted, not all central level media will be considered;
wherein news _ authority represents the report authority, (CC _ websites _ news)iThe number of news reports sent by the ith central level news media participating in the news reports on the same day is represented; (sites _ news)jNumber of stories, m, from the jth news platform targeted as important media that represents the current day's participation in news storiesiWeight corresponding to ith Central News media, njThe weight corresponding to the jth important media is determined according to the influence and the ranking on Alexa of each central level media and important media before calculation, and m and n are the number of the central level media and the number of the important media participating in the news report of the current day respectively;
(2) the report popularity is calculated based on the statistical information of news reports released by the news media and is equal to the weighted sum of the report popularity and the report interaction popularity; the report popularity is obtained by calculating the report quantity and the report rate, and the report interaction popularity is obtained by calculating the weighted sum of the forwarding number, the praise number and the comment number of the news reports, and the specific calculation mode is as follows:
wherein news _ heat represents news report heat, news _ numtodayIndicates the number of news reports of the day, max (news _ num)recent) Indicates the maximum number of recent news reports, max (news _ avg _ hour)todayIndicates the maximum number of news reports per hour of the day, i.e., the news reporting rate, max (news _ avg _ hour)recentIndicating the most recent news(ii) a reported rate;
news_inter_heat＝x*(news_like_heat)+y*(news_forw_heat)+z*(news_com_heat)
wherein, news _ inter _ heat represents the interactive heat of the news report, news _ like _ heat represents the like heat of the news report, news _ forew _ heat represents the forwarding heat of the news report, news _ com _ heat represents the comment heat of the news report, and x, y and z are weights obtained by an entropy weight method;
(3) the network public opinion popularity heat of the news media platform is obtained by normalizing the report authority and the report heat and then calculating a weighted sum, the weights of the report authority and the report heat are determined by an entropy weight method, and the specific calculation formula is as follows:
wherein, (news _ web _ heat)todayRepresents the internet public opinion popularity of the news media of the current day,
showing the normalized report authority,
Indicates the normalized reported heat,
And expressing the normalized report interaction heat.
10. The method for calculating and warning network public opinion popularity heat according to claim 4, wherein in the sixth step, the calculation formula of the current day's network public opinion heat is as follows:
xx_heattoday＝(news_web_heat)today+(social_heat)today
wherein xx heattodayAnd the network public opinion popularity of the current day is represented.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110693652.8A CN113609424A (en) | 2021-06-22 | 2021-06-22 | Computing and early warning system and method for network public sentiment popularity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110693652.8A CN113609424A (en) | 2021-06-22 | 2021-06-22 | Computing and early warning system and method for network public sentiment popularity |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113609424A true CN113609424A (en) | 2021-11-05 |
Family
ID=78303623
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110693652.8A Pending CN113609424A (en) | 2021-06-22 | 2021-06-22 | Computing and early warning system and method for network public sentiment popularity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113609424A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114187126A (en) * | 2021-11-08 | 2022-03-15 | 杭州萝卜智能技术有限公司 | Network public opinion management and control system based on data analysis |
| CN114615667A (en) * | 2022-05-12 | 2022-06-10 | 中国电子科技集团公司第三十研究所 | Method for suppressing network public opinion warning storm |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100191742A1 (en) * | 2009-01-27 | 2010-07-29 | Palo Alto Research Center Incorporated | System And Method For Managing User Attention By Detecting Hot And Cold Topics In Social Indexes |
| JP2011108053A (en) * | 2009-11-18 | 2011-06-02 | Nomura Research Institute Ltd | System for evaluating news article |
| US20120079020A1 (en) * | 2010-09-27 | 2012-03-29 | Korea Institute Of Science And Technology | Highlight providing system and method based on hot topic event detection |
| US20130173368A1 (en) * | 2011-09-29 | 2013-07-04 | Gregory Boutin | System and methods for popularity and influence indicators and commercial incentives based on object-related social network referrals |
| CN105447081A (en) * | 2015-11-04 | 2016-03-30 | 国云科技股份有限公司 | Cloud platform-oriented government affair and public opinion monitoring method |
| CN106021278A (en) * | 2016-04-27 | 2016-10-12 | 湖南蚁坊软件有限公司 | Method for analyzing territory heat degree index of micro blog event |
| CN107463686A (en) * | 2017-08-10 | 2017-12-12 | 深圳市腾讯计算机系统有限公司 | A kind of method and device of calculating network public sentiment temperature |
| US20180091611A1 (en) * | 2016-09-29 | 2018-03-29 | Thomson Reuters Global Resources Unlimited Company | Systems and methods for providing a social media knowledge base |
| CN109960756A (en) * | 2019-03-19 | 2019-07-02 | 国家计算机网络与信息安全管理中心 | Media event information inductive method |
| CN111581370A (en) * | 2020-04-20 | 2020-08-25 | 国家计算机网络与信息安全管理中心 | Network public opinion popularity heat degree evaluation method and device integrating multi-channel data sources |
| CN112434226A (en) * | 2020-12-15 | 2021-03-02 | 易研信息科技有限公司 | Network public opinion monitoring and early warning method |
| CN112711691A (en) * | 2021-01-08 | 2021-04-27 | 深圳市网联安瑞网络科技有限公司 | Network public opinion guide effect data information processing method, system, terminal and medium |
| CN112836137A (en) * | 2020-12-30 | 2021-05-25 | 深圳市网联安瑞网络科技有限公司 | Person network support degree calculation system and method, terminal, device, and storage medium |
-
2021
- 2021-06-22 CN CN202110693652.8A patent/CN113609424A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100191742A1 (en) * | 2009-01-27 | 2010-07-29 | Palo Alto Research Center Incorporated | System And Method For Managing User Attention By Detecting Hot And Cold Topics In Social Indexes |
| JP2011108053A (en) * | 2009-11-18 | 2011-06-02 | Nomura Research Institute Ltd | System for evaluating news article |
| US20120079020A1 (en) * | 2010-09-27 | 2012-03-29 | Korea Institute Of Science And Technology | Highlight providing system and method based on hot topic event detection |
| US20130173368A1 (en) * | 2011-09-29 | 2013-07-04 | Gregory Boutin | System and methods for popularity and influence indicators and commercial incentives based on object-related social network referrals |
| CN105447081A (en) * | 2015-11-04 | 2016-03-30 | 国云科技股份有限公司 | Cloud platform-oriented government affair and public opinion monitoring method |
| CN106021278A (en) * | 2016-04-27 | 2016-10-12 | 湖南蚁坊软件有限公司 | Method for analyzing territory heat degree index of micro blog event |
| US20180091611A1 (en) * | 2016-09-29 | 2018-03-29 | Thomson Reuters Global Resources Unlimited Company | Systems and methods for providing a social media knowledge base |
| CN107463686A (en) * | 2017-08-10 | 2017-12-12 | 深圳市腾讯计算机系统有限公司 | A kind of method and device of calculating network public sentiment temperature |
| CN109960756A (en) * | 2019-03-19 | 2019-07-02 | 国家计算机网络与信息安全管理中心 | Media event information inductive method |
| CN111581370A (en) * | 2020-04-20 | 2020-08-25 | 国家计算机网络与信息安全管理中心 | Network public opinion popularity heat degree evaluation method and device integrating multi-channel data sources |
| CN112434226A (en) * | 2020-12-15 | 2021-03-02 | 易研信息科技有限公司 | Network public opinion monitoring and early warning method |
| CN112836137A (en) * | 2020-12-30 | 2021-05-25 | 深圳市网联安瑞网络科技有限公司 | Person network support degree calculation system and method, terminal, device, and storage medium |
| CN112711691A (en) * | 2021-01-08 | 2021-04-27 | 深圳市网联安瑞网络科技有限公司 | Network public opinion guide effect data information processing method, system, terminal and medium |
Non-Patent Citations (3)
| Title |
|---|
| 丁晟春等: "基于关键词共现和社区发现的微博热点主题识别研究", 《现代情报》, vol. 38, no. 3, 31 March 2018 (2018-03-31), pages 10 - 18 * |
| 张寿华等: "网络舆情追踪中热点关键词的提取", 《河北大学学报（自然科学版）》, vol. 32, no. 3, 31 March 2012 (2012-03-31), pages 2 * |
| 徐勇;: "网络舆情事件演变的动力学建模及预警监测", 现代情报, no. 04, 15 April 2016 (2016-04-15), pages 3 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114187126A (en) * | 2021-11-08 | 2022-03-15 | 杭州萝卜智能技术有限公司 | Network public opinion management and control system based on data analysis |
| CN114187126B (en) * | 2021-11-08 | 2022-08-19 | 杭州萝卜智能技术有限公司 | Network public opinion management and control system based on data analysis |
| CN114615667A (en) * | 2022-05-12 | 2022-06-10 | 中国电子科技集团公司第三十研究所 | Method for suppressing network public opinion warning storm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7600017B2 (en) | System and method for scoring electronic messages | |
| Chan et al. | The internet and racial hate crime | |
| US7844483B2 (en) | System and method for predicting external events from electronic author activity | |
| US20120233253A1 (en) | Method and system for interacting and servicing users by orientation | |
| US20070219854A1 (en) | Document Examiner Comment System | |
| JP2008537817A (en) | Systems, methods, and computer program products for scoring items and determining predictor proficiency based on user sentiment | |
| CN113609424A (en) | Computing and early warning system and method for network public sentiment popularity | |
| JP2021099830A (en) | Method and global networking system for generation of global enterprise ranking in real-time based on globally acquired data | |
| Kroeber et al. | Still a glass ceiling? Tracing the limits to women’s representation in elected office | |
| CN112836137B (en) | Character network support degree computing system and method, terminal, equipment and storage medium | |
| CN117093762B (en) | Public opinion data evaluation analysis system and method | |
| Ghosh et al. | Effect of PACE on costs, nursing home admissions, and mortality: 2006-2011 | |
| CN113822596B (en) | Customer screening method based on big data | |
| CN113268976B (en) | Microblog-oriented topic influence assessment method | |
| Schroder | Attachment to place in New Zealand | |
| CN111768031A (en) | Method for predicting crowd gathering tendency based on ARMA algorithm | |
| Hesse | A coorientation study of Wisconsin state senators and their constituencies | |
| Hunter et al. | Family and social factors underlying the labour force status of Indigenous Australians | |
| Hwang et al. | Occupational self-selection and the gender wage gap: Evidence from Korea and United States | |
| Brewer et al. | Work from home, polarization, and new residential construction during COVID-19 | |
| Lu et al. | Predicting P2P lenders' decisions: the prospect theory approach | |
| Goldfarb | Learning in government programs and the usefulness of cost-benefit analysis: Lessons from manpower and urban renewal history | |
| Kocar et al. | Do we have to mix modes in probability-based online panel research to obtain more accurate results? | |
| Gafa et al. | Examining Poverty Dynamics in Ghana: Evidence from Longitudinal and Repeated Cross-Sectional Data | |
| Ni et al. | Fostering Entrepreneurship through Crowdfunding: What Drives Local Biases? |
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 |