CN114742281A - Public event network public opinion popularity prediction method based on grey model - Google Patents
Public event network public opinion popularity prediction method based on grey model Download PDFInfo
- Publication number
- CN114742281A CN114742281A CN202210277662.8A CN202210277662A CN114742281A CN 114742281 A CN114742281 A CN 114742281A CN 202210277662 A CN202210277662 A CN 202210277662A CN 114742281 A CN114742281 A CN 114742281A
- Authority
- CN
- China
- Prior art keywords
- data
- model
- method based
- prediction method
- public
- 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
Images
Classifications
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a public event network public opinion popularity prediction method based on a gray model, which comprises the following steps: s1, data collection operation is carried out by utilizing collector software; s2, preprocessing the collected data and calculating a heat trend value by using a model; s3, randomly selecting 500 pieces of data from the collected data to analyze, and judging the emotional tendency trend of the current day; and S4, putting the heat trend value into a prediction model for analysis, and predicting the trend of the heat trend. The invention realizes the prediction of public event network public opinion popularity by using the gray model.
Description
Technical Field
The invention is applied to the field of public opinion popularity prediction, and particularly relates to a public affair online public opinion popularity prediction method based on a gray model.
Background
The research focus of China on the popularity of the network public sentiment mainly focuses on multiple aspects of the construction of an evaluation index system of a hotspot event, the trend prediction of the popularity of a topic, the discovery of the hotspot topic, sentiment analysis and the like, and enterprises and research units determine the popularity evaluation index of the network public sentiment from the viewpoint of organically combining the qualitative and quantitative aspects. In the public sentiment popularity index system constructed in China, students mainly give rights around news events, netizens, clicks, post transfer amount and other indexes, and comprehensively evaluate the popularity of topics and predict the trend by using different technical means. How to accurately predict the public network opinion popularity of the public events becomes a technical problem to be solved in the field.
Disclosure of Invention
The invention aims to solve the technical problem of providing a public event network public opinion popularity prediction method based on a gray model aiming at the defects of the prior art.
In order to solve the technical problems, the public event network public opinion popularity prediction method based on the gray model comprises the following steps:
s1, data collection operation is carried out by utilizing collector software;
s2, preprocessing the collected data and calculating a heat trend value by using a model;
s3, randomly selecting 500 pieces of data from the collected data for analysis, and judging the emotional tendency trend of the day;
and S4, putting the heat trend value into a prediction model for analysis, and predicting the trend of the heat trend. As a possible implementation manner, further, the step S2 includes the following specific steps:
s21, removing duplicate of the collected data according to the Bo Wen, and filling the number '0' of the praise, the comment number and the forwarding number of the Bo Wen;
s22, carrying out weight calculation on the data after the de-weight filling;
s23, calculating the heat trend value of the day according to the weights of the praise number, the comment number and the forwarding number; using the calculation formula: h (i) ═ Ai*W1+Bi*W2+Ci*W3And (6) performing calculation.
As a possible implementation manner, further, the step S22 includes the following specific steps:
s221, carrying out standardization processing on the data by adopting a forward index formula;
s222, calculating the specific gravity of a certain index in a certain day by the following formula:
s224, carrying out information entropy redundancy calculation of a certain index, wherein the formula is as follows: dj=1-ej;
as a possible implementation manner, further, the step S3 includes the following specific steps: data preprocessing, emotion value calculation operation and experimental result output operation.
As a possible implementation manner, further, the data preprocessing operation specifically includes:
s31, removing stop words from the blog;
s32, Chinese word segmentation is carried out and classified into emotional words, negative words and degree adverbs.
As a possible implementation manner, further, the emotion value calculation operation specifically includes:
s33, finding out corresponding weight in corresponding emotion word stock;
and S34, calculating the emotion value according to the weights in the emotional words, the negative words and the degree adverbs in the blog text and judging the emotional tendency of the blog text.
As a possible implementation manner, further, the step S4 includes the following specific steps:
s41, accumulating the heat trend values for one time to generate a sequence;
s42, calculating an adjacent mean value generation sequence;
s43, constructing a data matrix and a data vector;
s44, calculating a development coefficient and an ash action amount;
s45, obtaining a time response function and predicting according to the function;
and S46, comparing the result obtained according to the time response function with the actual result, and if the result meets the model accuracy grade condition, passing the prediction, otherwise failing the prediction.
By adopting the technical scheme, the invention has the following beneficial effects: the method is used for constructing the microblog public opinion heat trend value model. And finally, calculating a public opinion popularity trend value according to the weight values, mainly calculating the popularity trend value of the microbo public opinion event, and providing a basis for constructing a GM model. And analyzing microblog public sentiment based on the BosonNLP sentiment dictionary. Chinese word segmentation is carried out on microblog messages and is divided into three categories of emotion words, negative words and degree adverbs, then emotion dictionaries are respectively matched to obtain corresponding emotion scores, and finally the microblog messages are divided into three categories of positive, negative and neutral according to the counted emotion score values. And predicting the popularity of the network public sentiment based on the GM model. The short-term prediction analysis is carried out on the trend value of the microblog public opinion popularity heat based on the GM model principle and the analysis of the model inspection method, and the result shows that the model precision has an obvious effect.
Drawings
The invention is described in further detail below with reference to the following figures and embodiments:
FIG. 1 is a flowchart illustrating step S3 according to the present invention;
FIG. 2 is a flowchart of step S4 according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be described clearly and completely in the following with reference to the accompanying drawings.
As shown in fig. 1-2, the present invention provides a public event internet public opinion heat prediction method based on a gray model, comprising the following steps:
s1, data collection operation is carried out by utilizing collector software; the preferred octopus collector is an octopus collector, and the octopus collector is a web crawler tool with simple operation and strong functions. According to the method, the 'octopus ocellatus collector' software is used, and the 'xx event' is used as a search keyword, original microblog blog text contents in the period from 4 month 1 to 4 month 30 in 2020 to the Xinlang microblog website are collected, and corresponding information such as the number of praise microblogs, the number of comments on microblogs, the number of forwarded microblogs and the like is collected, so that 604865 pieces of blog text are counted.
S2, preprocessing the collected data and calculating a heat trend value by using a model; the step S2 specifically includes:
s21, removing duplicate of the collected data according to the Bo Wen, and filling the number '0' of the praise, the comment number and the forwarding number of the Bo Wen;
s22, carrying out weight calculation on the data after the weight removal and filling; further, the specific step of step S22 includes:
s221, carrying out standardization processing on the data by adopting a forward index formula;
s222, calculating the specific gravity of a certain index in a certain day by the following formula:
s224, carrying out information entropy redundancy calculation of a certain index, wherein the formula is as follows: dj=1-ej;
s23, calculating the heat trend value of the day according to the weights of the praise number, the comment number and the forwarding number; using the calculation formula: h (i) ═ Ai*W1+Bi*W2+Ci*W3And (6) performing calculation.
S3, randomly selecting 500 pieces of data from the collected data for analysis, and judging the emotional tendency trend of the day; the method comprises the following specific steps: data preprocessing, emotion value calculation operation and experimental result output operation. Further, the data preprocessing operation specifically includes:
s31, removing stop words from the blog;
s32, performing Chinese word segmentation operation and classifying the words into sentiment words, negative words and degree adverbs.
Further, the emotion value calculation operation specifically includes:
s33, finding out corresponding weight in corresponding emotion word stock;
and S34, calculating the emotion value according to the weights in the emotional words, the negative words and the degree adverbs in the blog text and judging the emotional tendency of the blog text.
The specific operations include, for example: part of "i am happy and happy today" is that the word segmentation can be obtained directly by removing the characters that are not used: [ 'very', 'happy', 'very', 'happy' ]. Assume that the onset weight W is 1 and the emotion score is 0.
The first emotion word is happy, and because the emotion weight of the happy emotion is 1.4895, the emotion score is W and the emotion weight is 1.4895; then, a gap occurs between happy mood and the beginning of the next emotional word, the degree of which is 1.8, and thus the weight W is 1 × 1.8 — 1.8.
Then, the next emotion word, i.e. the word is open, the emotion weight is 2.6123, the weight W is 1.8, the emotion score is score +1.8 × 2.6123 is 6.1917, and the traversal is finished.
And S4, putting the heat trend value into a prediction model for analysis, and predicting the trend of the heat trend.
The method comprises the following specific steps:
s41, accumulating the heat trend values for one time to generate a sequence;
s42, calculating an adjacent mean value generation sequence;
s43, constructing a data matrix and a data vector;
s44, calculating a development coefficient and an ash action amount;
s45, obtaining a time response function and predicting according to the function;
and S46, comparing the result obtained according to the time response function with the actual result, and if the result meets the model precision grade condition, the prediction is passed, otherwise, the prediction fails.
The calculation in the operation process is as follows:
definition 4.1 setting X(0)As the original sequence:
X(0)=(x(0)(1),x(0)(2),...,x(0)(n))
X(1)is X(0)The sequence of one summation of (a), we note as 1-AGO sequence:
X(1)=(x(1)(1),x(1)(2),...,x(1)(n))
wherein the content of the first and second substances,Z(1)is X(1)The sequence generated by the close-to-average of (1):
Z(1)=(z(1)(2),z(1)(3),...,z(1)(n))
wherein
Definition 4.2 setting X(0)=(x(0)(1),x(0)(2),...,x(0)(n)), meterCalculating to obtain X(1)Is X(0)One accumulation of (a) to generate a sequence of:
X(1)=(x(1)(1),x(1)(2),...,x(1)(n))
then
x(0)(k)+ax(1)(k)=b
In the original form of the GM (1,1) model, a, b are the parameters to be distinguished.
Definition 4.3 setting X(0)=(x(0)(1),x(0)(2),...,x(0)(n)),X(1)Is X(0)A sequence generated by a single accumulation of Z(1)Is X(1)The sequence generated by the close-to-average of (1):
X(1)=(x1(1),x1(2),...,x1(n))
Z(1)=(z(1)(2),z(1)(3),...,z(1)(n))
then we give the basic form of the GM (1,1) model as follows:
x(0)(k)+az(1)(k)=b
definition 4.4 method according to matrix, we note the parameter columns asAnd recording the data series B and the data vector Y as follows:
the GM (1,1) model can be expressed asThe values of a and b are obtained by following the principle of unary linear regression, i.e. least squares, and there are
Definition 4.5 setting X(0)=(x(0)(1),x(0)(2),...,x(0)(n)),X(1)Is X(0)A sequence generated by a single accumulation of Z(1)Is X(1)The sequence generated by the close-to-average of (1):
X(1)=(x1(1),x1(2),...,x1(n))
Z(1)=(z(1)(2),z(1)(3),...,z(1)(n))
then we differentiate the form:
is referred to as x(0)(k)+az(1)(k) B whitening equation.
Then
(2) The time response function sequence of the GM (1,1) model is
Where k is 1, 2.
(3) Reduction number
Where k is 1, 2.
In the expression, the parameter a isIs composed ofThe parameter b is the gray effect amount, and reflects the information of the background value.
Let X(0)=(x(0)(1),x(0)(2),......,x(0)(n)) is an original sequence, and the simulated value of the original sequence obtained according to the GM (1,1) model is recorded as:
then the variance of the original sequence is:
wherein
Let the residual sequence E ═ (E (1), E (2),. E (n)), where
Then the variance of the residual is
In the above formula, i is 1, 2. Then we define the posterior difference ratio as C, then
And finally, judging the precision of the model according to the posterior difference ratio C in the table 1.
TABLE 1 model accuracy rating
Calculation of the specific embodiment:
(1) setting the trend value of microblog public opinion popularity at 1 day 4 month to 10 days 4 month as an original sequence X(0),
X(0)=[1167758,475332,785852,332917,529918,350850,174727, 336278,244218,241390]
(2) For original data X(0)Performing one-time accumulation to generate a sequence X(1)The following can be obtained:
X(1)=[1167758,1643090,2428942,2761859,3291777,3642627,38 17354,4153632,4397850,4639240]
(3) constructing a data series B and a data vector Y:
(4) determining parameters a and b by using a least square method:
[a,b]T=(BTB)-1BTY=[0.13,795795]T
(5) the time response sequence of the GM (1,1) model is:
(6) calculating a posterior difference ratio C for precision test:
the foregoing is directed to embodiments of the present invention, and equivalents, modifications, substitutions and variations may be made by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (7)
1. A public event network public opinion popularity prediction method based on a gray model is characterized by comprising the following steps:
s1, data collection operation is carried out by utilizing collector software;
s2, preprocessing the collected data and calculating a heat trend value by using a model;
s3, randomly selecting 500 pieces of data from the collected data to analyze, and judging the emotional tendency trend of the current day;
and S4, putting the heat trend value into a prediction model for analysis, and predicting the trend of the heat trend.
2. The public event internet public opinion popularity prediction method based on the gray model as claimed in claim 1, wherein: the step S2 includes the following steps:
s21, removing duplicate of the collected data according to the Bo Wen, and filling the number '0' of the praise, the comment number and the forwarding number of the Bo Wen;
s22, carrying out weight calculation on the data after the de-weight filling;
s23, calculating the heat trend value of the day according to the weights of the praise number, the comment number and the forwarding number; using the calculation formula: h (i) ═ Ai*W1+Bi*W2+Ci*W3And (6) performing calculation.
3. The public event internet public opinion popularity prediction method based on gray model as claimed in claim 2, characterized in that: the step S22 includes the following steps:
s221, carrying out standardization processing on the data by adopting a forward index formula;
s222, calculating the specific gravity of a certain index in a certain day by the following formula:
s224, carrying out information entropy redundancy calculation of a certain index, wherein the formula is as follows: dj=1-ej;
4. the public event network public opinion popularity prediction method based on the gray model as claimed in claim 1, characterized in that: the specific step of step S3 includes: data preprocessing, emotion value calculation operation and experimental result output operation.
5. The public event network public opinion popularity prediction method based on gray model as claimed in claim 4, wherein: the data preprocessing operation specifically comprises:
s31, removing stop words from the blog;
s32, Chinese word segmentation is carried out and classified into emotional words, negative words and degree adverbs.
6. The public event network public opinion popularity prediction method based on the gray model as claimed in claim 4, characterized in that: the emotion value calculation operation specifically includes:
s33, finding out corresponding weight in corresponding emotion word stock;
and S34, calculating the emotion value according to the weights in the emotional words, the negative words and the degree adverbs in the blog text and judging the emotional tendency of the blog text.
7. The public event internet public opinion popularity prediction method based on the gray model as claimed in claim 1, wherein: the step S4 includes the following steps:
s41, accumulating the heat trend values for one time to generate a sequence;
s42, calculating an adjacent mean value generation sequence;
s43, constructing a data matrix and a data vector;
s44, calculating a development coefficient and an ash action amount;
s45, obtaining a time response function and predicting according to the function;
and S46, comparing the result obtained according to the time response function with the actual result, and if the result meets the model precision grade condition, the prediction is passed, otherwise, the prediction fails.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210277662.8A CN114742281A (en) | 2022-03-21 | 2022-03-21 | Public event network public opinion popularity prediction method based on grey model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210277662.8A CN114742281A (en) | 2022-03-21 | 2022-03-21 | Public event network public opinion popularity prediction method based on grey model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114742281A true CN114742281A (en) | 2022-07-12 |
Family
ID=82277314
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210277662.8A Pending CN114742281A (en) | 2022-03-21 | 2022-03-21 | Public event network public opinion popularity prediction method based on grey model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114742281A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117076751A (en) * | 2023-10-10 | 2023-11-17 | 西安康奈网络科技有限公司 | Public opinion event development trend judging system based on multidimensional feature analysis |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110728140A (en) * | 2018-06-30 | 2020-01-24 | 天津大学 | Emotion analysis and theme feature-based emergent event public opinion evolution analysis method |
CN111460158A (en) * | 2020-04-01 | 2020-07-28 | 安徽理工大学 | Microblog topic public emotion prediction method based on emotion analysis |
CN113779382A (en) * | 2021-08-19 | 2021-12-10 | 三江学院 | Network public opinion prediction method based on microblog data |
-
2022
- 2022-03-21 CN CN202210277662.8A patent/CN114742281A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110728140A (en) * | 2018-06-30 | 2020-01-24 | 天津大学 | Emotion analysis and theme feature-based emergent event public opinion evolution analysis method |
CN111460158A (en) * | 2020-04-01 | 2020-07-28 | 安徽理工大学 | Microblog topic public emotion prediction method based on emotion analysis |
CN113779382A (en) * | 2021-08-19 | 2021-12-10 | 三江学院 | Network public opinion prediction method based on microblog data |
Non-Patent Citations (3)
Title |
---|
彭艺 等: "一种基于马尔可夫链的微信舆情热度预测模型" * |
曾子明 等: "基于BP神经网络的突发传染病舆情热度趋势预测模型研究" * |
李文娟 等: "基于灰色模型的微博舆情预测研究" * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117076751A (en) * | 2023-10-10 | 2023-11-17 | 西安康奈网络科技有限公司 | Public opinion event development trend judging system based on multidimensional feature analysis |
CN117076751B (en) * | 2023-10-10 | 2024-01-16 | 西安康奈网络科技有限公司 | Public opinion event development trend judging system based on multidimensional feature analysis |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Haselmayer et al. | Sentiment analysis of political communication: Combining a dictionary approach with crowdcoding | |
CA3098802C (en) | Systems and methods for generating a contextually and conversationally correct response to a query | |
Sehgal et al. | Sops: stock prediction using web sentiment | |
CN111967761A (en) | Monitoring and early warning method and device based on knowledge graph and electronic equipment | |
CN112035658A (en) | Enterprise public opinion monitoring method based on deep learning | |
CN113378565A (en) | Event analysis method, device and equipment for multi-source data fusion and storage medium | |
Lina et al. | Fake news detection in the urdu language using CharCNN-RoBERTa | |
Tang et al. | Evaluation of Chinese sentiment analysis APIs based on online reviews | |
CN115329085A (en) | Social robot classification method and system | |
Zhang et al. | A hybrid neural network approach for fine-grained emotion classification and computing | |
CN114742281A (en) | Public event network public opinion popularity prediction method based on grey model | |
Pai et al. | Real-time Twitter sentiment analytics and visualization using Vader | |
Song et al. | Research on Kano model based on online comment data mining | |
Jawad et al. | Combination Of Convolution Neural Networks And Deep Neural Networks For Fake News Detection | |
CN111859955A (en) | Public opinion data analysis model based on deep learning | |
CN109726938A (en) | A kind of students ' thinking political affairs situation method for early warning based on deep learning | |
CN111598691B (en) | Method, system and device for evaluating default risk of credit/debt main body | |
Hajare et al. | A machine learning pipeline to examine political bias with congressional speeches | |
Soe et al. | Domain oriented aspect detection for student feedback system | |
KR20090126862A (en) | System and method for analyzing emotional information from natural language sentence, and medium for storaging program for the same | |
CN114511345A (en) | Sales prediction method based on policy-public opinion-purchase double-order deep learning | |
Nguyen et al. | Opinion spam recognition method for online reviews using ontological features | |
AlKhatib et al. | Analysing the Sentiments of Opinion Leaders in Relation to Smart Cities’ Major Events | |
Unnikrishnan et al. | A Literature Review of Sentiment Evolution | |
Erfina et al. | Indonesian Analysis Sentiment on Non Fungible Token (NFT) |
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 |