CN112395513A - Public opinion transmission power analysis method - Google Patents
Public opinion transmission power analysis method Download PDFInfo
- Publication number
- CN112395513A CN112395513A CN202011375251.XA CN202011375251A CN112395513A CN 112395513 A CN112395513 A CN 112395513A CN 202011375251 A CN202011375251 A CN 202011375251A CN 112395513 A CN112395513 A CN 112395513A
- Authority
- CN
- China
- Prior art keywords
- information
- analysis
- emotion
- data
- public opinion
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 81
- 230000005540 biological transmission Effects 0.000 title claims abstract description 18
- 230000008451 emotion Effects 0.000 claims abstract description 44
- 238000000034 method Methods 0.000 claims abstract description 29
- 230000007935 neutral effect Effects 0.000 claims abstract description 7
- 238000012216 screening Methods 0.000 claims abstract description 6
- 238000012549 training Methods 0.000 claims description 16
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 5
- 230000009193 crawling Effects 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000013589 supplement Substances 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
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- 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
- G06F16/955—Retrieval from the web using information identifiers, e.g. uniform resource locators [URL]
- G06F16/9566—URL specific, e.g. using aliases, detecting broken or misspelled links
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention belongs to the technical field of information analysis, and particularly relates to a public opinion transmission power analysis method, which comprises the following steps: s1, collecting data information from the appointed information entrance; s2, standardizing the collected data information; s3, performing emotion analysis on the standardized data information by using a preset information emotion model, wherein the emotion analysis result comprises a positive attribute, a neutral attribute and a negative attribute; s4, storing the emotion analysis result into a database, and counting the data result after emotion analysis; s5, screening out data results with negative attributes, and carrying out propagation path analysis and key propagation point analysis; and S6, displaying the statistical result and the analysis result with the negative attribute. By using the method, the network public sentiment can be monitored and analyzed accurately and quickly in real time, the propagation chain and key propagation nodes of the negative attribute public sentiment can be found quickly, and corresponding measures can be taken timely and accurately.
Description
Technical Field
The invention belongs to the technical field of information analysis, and particularly relates to a public opinion transmission power analysis method.
Background
With the development of networks, especially the rapid development and iteration of new media technologies, the boundaries of global information become more and more fuzzy, and users can learn the world through various channels. The online public opinion is active to an unprecedented extent, and can be formed immediately no matter in domestic or international important events.
The new media brings great monitoring pressure while bringing convenience. The network friend expresses the view point through the network, and can cause huge public opinion influence in a very short time through modes of forwarding, transshipping and the like. If public opinion is negative, negative emotion spreading can be caused, and if the treatment is not timely, adverse effects can be caused.
Therefore, a public opinion transmission analysis method is needed, which can monitor and analyze network public opinions accurately and quickly in real time, so that related personnel can know the real-time situation of the public opinions and can respond in time when intervention is needed.
Disclosure of Invention
The invention aims to provide a public opinion transmission analysis method, which can be used for monitoring and analyzing network public opinions accurately and quickly in real time, so that related personnel can know the real-time situation of the public opinions and can respond in time when intervention is needed.
The basic scheme provided by the invention is as follows:
a public opinion transmission power analysis method comprises the following steps:
s1, collecting data information from the appointed information entrance;
s2, standardizing the collected data information;
s3, performing emotion analysis on the standardized data information by using a preset information emotion model, wherein the emotion analysis result comprises a positive attribute, a neutral attribute and a negative attribute;
s4, storing the emotion analysis result into a database, and counting the data result after emotion analysis;
s5, screening out data results with negative attributes, and carrying out propagation path analysis and key propagation point analysis;
and S6, displaying the statistical result and the analysis result with the negative attribute.
Basic scheme theory of operation and beneficial effect:
by using the method, data information is acquired from a designated information inlet, and after the acquired information is subjected to standardization processing, the standardized data information is subjected to emotion analysis by using a preset information emotion model. And then, counting the emotion analysis result, and screening out the result with negative attribute data.
When the analysis results with negative attributes are found, the data results with negative attributes are screened out and subjected to path propagation analysis and key propagation point analysis.
Through path propagation analysis, the propagation paths can be mined layer by layer, the specific propagation process of the negative attribute information is known, and the negative attribute information is convenient to process from the source. Through key propagation point analysis, key propagation nodes, namely key users with a large number of forwarding/replying/postings, can be known, and key breakthrough objects can be found conveniently when public opinion propagation is controlled.
By using the method, the network public sentiment can be monitored and analyzed accurately and quickly in real time, so that related personnel can know the real-time situation of the public sentiment. And when negative public sentiment exists, the propagation chain and the key propagation node can be quickly found, and corresponding measures can be timely and accurately taken when intervention is needed.
Further, in S1, the step of collecting data includes:
s11, crawling information on the initial page by taking a preset data acquisition page as an entry point, and forming effective structured data by using crawled contents;
s12, forming a new URL by the UID of the user and the URL of the current page, and storing the new URL into the data acquisition list entry;
and S13, storing the crawled data in a local database.
By the arrangement, the current page content is convenient to crawl, and subsequent data updating is convenient.
Further, in S2, the normalization process includes word segmentation, text classification, and clustering.
Further, in S3, the preset information emotion model training step includes:
s311, capturing information content for training;
s312, extracting the characteristics of the information for training;
and S313, training and establishing an information emotion model according to the information features for training and the machine learning algorithm module.
Further, in S3, the emotion analyzing step includes:
s321, capturing information content to be identified;
s322, extracting the characteristics of the information to be identified;
and S323, judging whether the information belongs to a positive side, a neutral side or a negative side by using a preset information emotion model according to the characteristics of the information to be identified.
Further, in S312 and S322, the techniques used in feature extraction both include an intelligent word segmentation technique and a text similarity technique in natural language analysis.
Further, in S1, the information entry for collecting data includes a website, a client, a microblog, a headline number, and a wechat public number.
Data information is collected through a whole network of multi-level channels, omnibearing intelligent analysis can be carried out on public sentiment from multiple dimensions, and the analysis result is more objective and comprehensive.
Further, in S1, data information is acquired at a preset frequency.
The frequency of data information acquisition can be set according to specific conditions, and system resources are reasonably utilized.
Further, in S6, the display mode is a visual icon display.
The statistical result can be understood more clearly.
Drawings
Fig. 1 is a flowchart of a public opinion transmission power analysis method according to a first embodiment of the present invention.
Detailed Description
The following is further detailed by way of specific embodiments:
example one
As shown in fig. 1, a public opinion transmission power analysis method includes:
s1, collecting data information from the appointed information entrance according to the preset frequency; specifically, the information entry for collecting data includes a website, a client, a microblog, a headline number and a wechat public number.
Specifically, the method comprises the following steps: s11, crawling information on the initial page by taking a preset data acquisition page as an entry point, and forming effective structured data by using crawled contents; s12, forming a new URL by the UID of the user and the URL of the current page, and storing the new URL into the data acquisition list entry; and S13, storing the crawled data in a local database.
S2, standardizing the collected data information; in this embodiment, the normalization process includes word segmentation, text classification, and clustering.
And S3, performing emotion analysis on the standardized data information by using a preset information emotion model, wherein the emotion analysis result comprises a positive attribute, a neutral attribute and a negative attribute.
The preset information emotion model training method comprises the following steps: s311, capturing information content for training; s312, extracting the characteristics of the information for training; and S313, training and establishing an information emotion model according to the information features for training and the machine learning algorithm module.
The emotion analysis method comprises the following steps: s321, capturing information content to be identified; s322, extracting the characteristics of the information to be identified; and S323, judging whether the information belongs to a positive side, a neutral side or a negative side by using a preset information emotion model according to the characteristics of the information to be identified.
In S312 and S322, the techniques used in feature extraction include intelligent word segmentation and text similarity in natural language analysis.
S4, storing the emotion analysis result into a database, and counting the data result after emotion analysis;
s5, screening out data results with negative attributes, and carrying out propagation path analysis and key propagation point analysis;
and S6, displaying the statistical result and the analysis result with the negative attribute. In this embodiment, the display mode is a visual icon display.
The specific implementation process is as follows:
by using the method, the information entry for collecting data comprises a website, a client, a microblog, a headline number and a WeChat public number. Data information is collected through a whole network of multi-level channels, omnibearing intelligent analysis can be carried out on public sentiment from multiple dimensions, and the analysis result is more objective and comprehensive.
Data information is collected from a designated information inlet, and after the collected information is subjected to standardization processing, emotion analysis is performed on the standardized data information by using a preset information emotion model. And then, counting the emotion analysis result, and screening out the result with negative attribute data.
When the analysis results with negative attributes are found, the data results with negative attributes are screened out and subjected to path propagation analysis and key propagation point analysis.
Through path propagation analysis, the propagation paths can be mined layer by layer, the specific propagation process of the negative attribute information is known, and the negative attribute information is convenient to process from the source. Through key propagation point analysis, key propagation nodes, namely key users with a large number of forwarding/replying/postings, can be known, and key breakthrough objects can be found conveniently when public opinion propagation is controlled.
In practical use, the method can carry out all-around acquisition on the information ports of websites, clients, microblogs, headline numbers, WeChat public numbers, and the like, no dead angle is left, and data are acquired in a directional acquisition mode and a non-directional supplement mode, wherein the data are about twenty-ten thousand + every day. Distributed cluster collection, wherein 200 websites and 1000 collectors are supported by a single collector to be parallel, the websites, the forums and the like are polled once in 3 minutes at the fastest speed, and microblog information is pushed in real time. 7, continuously collecting for 24 hours by a plurality of cloud servers; the automatic clustering is 1000/min, and the keyword search response time of 1 hundred million articles is 0.03 second. The template and script engine realize metadata extraction; the information extraction accuracy rate reaches more than 99%. Automatically analyzing the page hierarchical relation and acquiring the deepest content; forum postings, microblog comments and website comments can be collected.
By using the method, the network public sentiment can be monitored and analyzed accurately and quickly in real time, so that related personnel can know the real-time situation of the public sentiment. And when negative public sentiment exists, the propagation chain and the key propagation node can be quickly found, and corresponding measures can be timely and accurately taken when intervention is needed.
Example two
Because some internet users do not directly speak the words they want to say in order to display their expression ability and skill when they express their own opinions, especially when they have negative emotions, they hide the contents they want to express in the expressed contents in a hidden way, and the expressed contents are analyzed linguistically without negative contents. For these contents, the technical solution in the first embodiment cannot find the contents that the publisher really wants to express. These users usually have a large audience group and influence due to their expression and skill. Although a negative propagation effect is produced, the source of the negative propagation is not found using the technical solution of the embodiment.
In the current art, there is a case where an analysis of the ironic content is performed using a linguistic analysis technique such as NLP. However, the language habits of the internet are updated very quickly, and a trained analysis model is ineffective after a long time, so that the training is required to be put into practice again. Firstly, the efficiency is low, secondly, the timeliness is difficult to guarantee, and many ironic contents are difficult to process in time because the ironic contents are not recognized.
The above-mentioned problems are not solved, and unlike the first embodiment, in the present embodiment,
further comprising:
s11, when the collected data information includes the message information, classifying the data information into published information and message information, and associating;
s41, comparing and analyzing the emotion analysis data of the published information and the emotion analysis data of the associated message information, and marking the published content as suspected connotation content if the comparison and analysis result shows that the published information is normal but the associated message information is abnormal;
and S61, displaying the suspected content.
The specific implementation process is as follows:
when the internet user hides his negative viewpoint in the expressed contents in a hidden manner in order to show his expressive ability and skill, the normal analysis method treats the contents as normal contents, and even if the existing NLP system is adopted to perform ironic analysis, it is difficult to guarantee timeliness.
By using the method, when the collected data information comprises the message information, the data information is classified into the published information and the message information and is associated. And then comparing and analyzing the emotion analysis data of the published information and the emotion analysis data of the associated message information, and marking the published content as the suspected connotation content if the comparison and analysis result shows that the published information is normal but the associated message information is abnormal. Thus, although it is not possible to identify a meaningful number of articles with a negative opinion, it is sufficient to identify and mark the vast majority of such articles. Then, the suspected content is displayed and checked and identified by staff.
Although due to the internet user's jumpy thinking, with this approach, it sometimes happens that it is recognized as a connotative article because of normal taunt comments. However, such cases are only a few.
In addition, when the negative viewpoint has a certain propagation, compared with the existing other methods, the method can also quickly check and lock the key propagation nodes for propagating the content articles.
The foregoing is merely an example of the present invention, and common general knowledge in the field of known specific structures and characteristics is not described herein in any greater extent than that known in the art at the filing date or prior to the priority date of the application, so that those skilled in the art can now appreciate that all of the above-described techniques in this field and have the ability to apply routine experimentation before this date can be combined with one or more of the present teachings to complete and implement the present invention, and that certain typical known structures or known methods do not pose any impediments to the implementation of the present invention by those skilled in the art. It should be noted that, for those skilled in the art, without departing from the structure of the present invention, several changes and modifications can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicability of the patent. The scope of the claims of the present application shall be determined by the contents of the claims, and the description of the embodiments and the like in the specification shall be used to explain the contents of the claims.
Claims (9)
1. A public opinion transmission power analysis method is characterized by comprising the following steps:
s1, collecting data information from the appointed information entrance;
s2, standardizing the collected data information;
s3, performing emotion analysis on the standardized data information by using a preset information emotion model, wherein the emotion analysis result comprises a positive attribute, a neutral attribute and a negative attribute;
s4, storing the emotion analysis result into a database, and counting the data result after emotion analysis;
s5, screening out data results with negative attributes, and carrying out propagation path analysis and key propagation point analysis;
and S6, displaying the statistical result and the analysis result with the negative attribute.
2. The public opinion transmission analysis method according to claim 1, characterized in that: in S1, the step of collecting data includes:
s11, crawling information on the initial page by taking a preset data acquisition page as an entry point, and forming effective structured data by using crawled contents;
s12, forming a new URL by the UID of the user and the URL of the current page, and storing the new URL into the data acquisition list entry;
and S13, storing the crawled data in a local database.
3. The public opinion transmission analysis method according to claim 2, characterized in that: in S2, the normalization process includes word segmentation, text classification, and clustering.
4. The public opinion transmission analysis method according to claim 3, characterized in that: in S3, the preset information emotion model training step includes:
s311, capturing information content for training;
s312, extracting the characteristics of the information for training;
and S313, training and establishing an information emotion model according to the information features for training and the machine learning algorithm module.
5. The public opinion transmission analysis method according to claim 4, characterized in that: in S3, the emotion analyzing step includes:
s321, capturing information content to be identified;
s322, extracting the characteristics of the information to be identified;
and S323, judging whether the information belongs to a positive side, a neutral side or a negative side by using a preset information emotion model according to the characteristics of the information to be identified.
6. The public opinion transmission analysis method according to claim 5, characterized in that: in S312 and S322, the techniques used in feature extraction include intelligent word segmentation and text similarity in natural language analysis.
7. The public opinion transmission analysis method according to claim 6, characterized in that: in S1, the information entry for collecting data includes a website, a client, a microblog, a headline number, and a wechat public number.
8. The public opinion transmission analysis method according to claim 7, characterized in that: in S1, data information is acquired at a predetermined frequency.
9. The public opinion transmission analysis method according to claim 8, characterized in that: in S6, the display mode is visual icon display.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011375251.XA CN112395513A (en) | 2020-11-30 | 2020-11-30 | Public opinion transmission power analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011375251.XA CN112395513A (en) | 2020-11-30 | 2020-11-30 | Public opinion transmission power analysis method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112395513A true CN112395513A (en) | 2021-02-23 |
Family
ID=74605697
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011375251.XA Pending CN112395513A (en) | 2020-11-30 | 2020-11-30 | Public opinion transmission power analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112395513A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113111269A (en) * | 2021-05-10 | 2021-07-13 | 网易（杭州）网络有限公司 | Data processing method and device, computer readable storage medium and electronic equipment |
| CN114547167A (en) * | 2022-01-27 | 2022-05-27 | 启明信息技术股份有限公司 | Automobile public opinion sentiment analysis method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109145216A (en) * | 2018-08-29 | 2019-01-04 | 中国平安保险(集团)股份有限公司 | Network public-opinion monitoring method, device and storage medium |
| CN110533212A (en) * | 2019-07-04 | 2019-12-03 | 西安理工大学 | Urban waterlogging public sentiment monitoring and pre-alarming method based on big data |
| CN111538888A (en) * | 2020-06-05 | 2020-08-14 | 国网山东省电力公司检修公司 | Network public opinion intensity evolution analysis system based on active monitoring engine and big data |
-
2020
- 2020-11-30 CN CN202011375251.XA patent/CN112395513A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109145216A (en) * | 2018-08-29 | 2019-01-04 | 中国平安保险(集团)股份有限公司 | Network public-opinion monitoring method, device and storage medium |
| CN110533212A (en) * | 2019-07-04 | 2019-12-03 | 西安理工大学 | Urban waterlogging public sentiment monitoring and pre-alarming method based on big data |
| CN111538888A (en) * | 2020-06-05 | 2020-08-14 | 国网山东省电力公司检修公司 | Network public opinion intensity evolution analysis system based on active monitoring engine and big data |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113111269A (en) * | 2021-05-10 | 2021-07-13 | 网易（杭州）网络有限公司 | Data processing method and device, computer readable storage medium and electronic equipment |
| CN114547167A (en) * | 2022-01-27 | 2022-05-27 | 启明信息技术股份有限公司 | Automobile public opinion sentiment analysis method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109829089B (en) | Social network user anomaly detection method and system based on associated graph | |
| WO2018036239A1 (en) | Method, apparatus and system for monitoring internet media events based on industry knowledge mapping database | |
| CN108776671A (en) | A kind of network public sentiment monitoring system and method | |
| CN106815307A (en) | Public Culture knowledge mapping platform and its use method | |
| CN110533212A (en) | Urban waterlogging public sentiment monitoring and pre-alarming method based on big data | |
| Whitney et al. | Don’t want to get caught? don’t say it: The use of emojis in online human sex trafficking ads | |
| CN112395513A (en) | Public opinion transmission power analysis method | |
| CN110717089A (en) | User behavior analysis system and method based on weblog | |
| Memon et al. | Harvesting covert networks: a case study of the iMiner database | |
| Hu | Big data analysis of criminal investigations | |
| KR101665649B1 (en) | System for analyzing social media data and method for analyzing social media data using the same | |
| US10990881B1 (en) | Predictive analytics using sentence data model | |
| Phillips et al. | Extracting social structure from darkweb forums | |
| Yang et al. | Public cultural knowledge graph platform | |
| CN111026940A (en) | Network public opinion and risk information monitoring system and electronic equipment for power grid electromagnetic environment | |
| CN116521729A (en) | Information classification searching method and device based on elastic search | |
| Bharathi et al. | A supervised learning approach for criminal identification using similarity measures and K-Medoids clustering | |
| Yin et al. | Research of integrated algorithm establishment of a spam detection system | |
| Qureshi et al. | Detecting social polarization and radicalization | |
| Abele | Linked data profiling: Identifying the domain of datasets based on data content and metadata | |
| ur Rahman et al. | Centralized vulnerability database for organization specific automated vulnerabilities discovery and supervision | |
| Lee et al. | Exploiting online social data in ontology learning for event tracking and emergency response | |
| Levshun et al. | Active learning approach for inappropriate information classification in social networks | |
| Yuan et al. | OPO: Online public opinion analysis system over text streams | |
| Chen et al. | Research on Network Intelligent Finance Data Mining and Analysis System and Quantitative Model |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210223 |