CN117494897A - Single public opinion event development tendency judging method - Google Patents
Single public opinion event development tendency judging method Download PDFInfo
- Publication number
- CN117494897A CN117494897A CN202311512337.6A CN202311512337A CN117494897A CN 117494897 A CN117494897 A CN 117494897A CN 202311512337 A CN202311512337 A CN 202311512337A CN 117494897 A CN117494897 A CN 117494897A
- Authority
- CN
- China
- Prior art keywords
- public opinion
- development
- event
- viewpoint
- day
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000011161 development Methods 0.000 title claims abstract description 316
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000011156 evaluation Methods 0.000 claims abstract description 35
- 238000012937 correction Methods 0.000 claims abstract description 22
- 238000004458 analytical method Methods 0.000 claims abstract description 21
- 230000008859 change Effects 0.000 claims description 21
- 230000007935 neutral effect Effects 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 8
- CLOMYZFHNHFSIQ-UHFFFAOYSA-N clonixin Chemical compound CC1=C(Cl)C=CC=C1NC1=NC=CC=C1C(O)=O CLOMYZFHNHFSIQ-UHFFFAOYSA-N 0.000 claims description 6
- 230000003993 interaction Effects 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 101000912561 Bos taurus Fibrinogen gamma-B chain Proteins 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 claims description 3
- 230000002452 interceptive effect Effects 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 239000000463 material Substances 0.000 claims description 2
- 238000012790 confirmation Methods 0.000 abstract description 3
- 230000010485 coping Effects 0.000 abstract description 3
- 230000007812 deficiency Effects 0.000 abstract 2
- 230000007547 defect Effects 0.000 description 7
- 230000004888 barrier function Effects 0.000 description 6
- 238000001914 filtration Methods 0.000 description 4
- 230000008451 emotion Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 230000000903 blocking effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013210 evaluation model Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
Abstract
The invention belongs to the technical field of public opinion event development tendency judgment, and particularly discloses a single public opinion event development tendency judgment method, which comprises the following steps: extracting the content of the monitored target public opinion event, counting the current development days of the target public opinion event, and extracting the development data of the target public opinion event in each development day; carrying out similar historical public opinion event confirmation; analyzing the public opinion positive trend degree of the target public opinion event; setting a public opinion positive trend evaluation correction factor and correcting; judging the development tendency of the target public opinion time, and feeding back; the invention realizes the multi-aspect judgment of the public opinion event, effectively solves the problem that the current public opinion event development tendency judgment still has a certain deficiency, fully considers the timeliness of the public opinion event, makes up the deficiency in the current public opinion event development variability analysis, and is convenient for public opinion related management staff to grasp the opportunity of coping with the public opinion event and properly deal with the potential crisis.
Description
Technical Field
The invention belongs to the technical field of public opinion event development tendency judgment, and particularly relates to a single public opinion event development tendency judgment method.
Background
Single public opinion event development tendency analysis is a process of evaluating and predicting how a particular public opinion event develops. And public opinion events may have a significant impact on parties such as enterprises, organizations, or governments. The importance and necessity of analysis of the tendency of public opinion event development are highlighted.
The current single public opinion event development tendency analysis mainly carries out development tendency judgment according to related comments and forwarding data from public opinion event disclosure, and obviously, the current public opinion event development tendency judgment has certain defects, and the specific aspects are as follows: 1. the advanced dynamic analysis is not carried out on the public opinion development data, so that the public opinion event development tendency judgment is shallow, and the public opinion event development tendency judgment error is large, so that the authenticity and reliability of the public opinion event development tendency judgment result cannot be ensured, and the blocking timeliness and blocking effect of negative public opinion development cannot be improved.
2. The data utilization rate of the historical related public opinion event is not high, the current public opinion event development tendency judgment basis is relatively fixed, the rationality and standardization of the public opinion event development tendency judgment cannot be ensured, and the referential of the public opinion event development tendency judgment cannot be improved.
3. The timeliness of the public opinion event is considered insufficiently, so that the analysis of the development variability of the public opinion event is insufficient, wrong tendency judgment is easy to generate, meanwhile, the possibility of missing the opportunity of the public opinion event or failing to properly handle the potential crisis is also possibly caused, the public opinion related manager is inconvenient to adjust strategies and actions in time, and the public attention point, the demand and the emotion change cannot be insight, so that correct information and response cannot be provided pertinently, and the public opinion guidance cannot be guided effectively in the follow-up.
Disclosure of Invention
In view of this, in order to solve the above-mentioned problems in the background art, a single public opinion event development tendency judging method is proposed.
The aim of the invention can be achieved by the following technical scheme: the invention provides a single public opinion event development tendency judging method, which comprises the following steps: step 1, extracting basic monitoring data of public opinion events: and extracting the content and the current development days of the monitored target public opinion event.
Step 2, extracting public opinion event development data: and extracting development data of the monitored target public opinion event in each development day.
Step 3, public opinion rating reference data analysis: and extracting the content, the number of development days and the development data in each development day of each history public opinion event from the public opinion information base, and confirming each similar history public opinion event.
And 4, public opinion event development prediction analysis: public opinion positive trend lambda for analyzing target public opinion event Positive direction 。
Step 5, public opinion event development prediction correction: setting a public opinion positive trend evaluation correction factor eta so as to predict the estimated public opinion positive trend lambda 'of the target public opinion event' Positive direction ，λ′ Positive direction ＝λ Positive direction *(1-η)。
Step 6, public opinion event development tendency judgment: will be lambda' Positive direction And importing the target public opinion event into a development tendency judgment model, and outputting the development tendency of the target public opinion event.
Step 7, public opinion event development tendency feedback: and feeding back the development tendency of the target public opinion event to relevant management and control personnel of the target public opinion event.
Preferably, the development data includes opinion publishing data and interaction data.
The viewpoint posting data comprises the number of posting viewpoint users, the attribute of each posting viewpoint user, an ID account number, a fan amount and posting viewpoint contents, and the interactive data comprises the reading amount, the praise amount, the forwarding amount, the number of comment users, the comment contents of each comment user, the number of reply users and the reply contents of each reply user corresponding to the posting viewpoint contents of each posting viewpoint user.
Preferably, the identifying each similar historical public opinion event includes: keyword recognition is respectively carried out on the contents of the target public opinion event and the historical public opinion event through keyword recognition technology, so that keywords corresponding to the target public opinion event and the historical public opinion event are obtained, and then the content similarity χ of the historical public opinion event is counted i I represents a historical public opinion event number, i=1, 2.
The current development day number of the target public opinion event is recorded as R 0 。
Sequencing the development days corresponding to the historical public opinion events according to time sequence, and extracting R before ranking from the development data in the development days corresponding to the historical public opinion events 0 Development data of each development day, and statistics of development similarity beta of each public opinion event i 。
The comprehensive public opinion similarity gamma i corresponding to each public opinion event of the history is counted,χ 'and β' are respectively the content similarity and the development similarity of the set reference, a1 and a2 are respectively the content similarity and the development similarity corresponding to the comprehensive public opinion similarity evaluation duty ratio weight, and σ is the set public opinion similarity evaluation correction factor>Representing rounding down symbols.
Will gamma i And comparing the evaluation similarity gamma 'of the set similar public opinion events, and screening out each historical public opinion event larger than gamma' as each similar historical public opinion event.
Preferably, the statistics of the development similarity of each public opinion event includes: the number of the development days of each public opinion event is recorded as R' i 。
If R 'is' i ≤R 0 The i-th historical public opinion event is marked as a non-similar public opinion event, and the development similarity of the non-similar public opinion event is marked as
If R 'is' i ＞R 0 And the ith historical public opinion event is recorded as a reference public opinion event.
Numbering each development day of the target public opinion event, and extracting the number M of users publishing opinion from the development data in each development day 0 t T represents a day of development number, t=1, 2.
Will refer to public opinion event corresponding top R 0 The development data of each development day is used as reference development data of target public opinion event corresponding to each development day, and the number M of users publishing opinion is extracted from the reference development data 1 t 。
Extracting attributes of users from the published views from the development data in each development day, and counting the user distribution richness F of each development day t Mass ratio k Group of t And a media duty k Media (media) t 。
According to reference development data in each development day, according to F t 、k Group of t 、k Media (media) t Statistical means of (c) and (d) statistics of user distribution richness F 'of reference in each development day' t Mass ratioAnd a media duty k' Media (media) 。
The similarity XY of participating users with reference to public opinion is counted,
，ΔM、ΔF、Δk group of 、Δk Media (media) The method comprises the steps of respectively setting the difference of the number of the participating users permitted by the similar public opinion event, the difference of the abundance of the user distribution, the difference of the mass ratio and the difference of the media ratio.
Extracting the reading quantity, the praise quantity, the forwarding quantity and the comment user number of the corresponding publishing viewpoint content of each publishing viewpoint user from the development data in each development day, and counting the development heat tendency degree in each development day, wherein the trend degree is marked as phi t And similarly counting the trend phi 'of the reference development heat in each development day' t 。
The development heat similarity FD of the reference public opinion event is counted, and then the development similarity psi of the reference public opinion event is counted,XY ', FD' are the similarity threshold and development of the participating users for setting reference respectivelyA heat similarity threshold value to obtain the development similarity beta of each public opinion event of the history i ，β i The value is +.>Or a combination of the above-mentioned materials,
preferably, the analyzing the public opinion positive tendency of the target public opinion event includes: and marking the publishing viewpoint user, the comment user and the reply user as a first viewpoint user, a second viewpoint user and a third viewpoint user respectively.
And extracting comment contents of each secondary viewpoint user and reply contents of each tertiary viewpoint user under the corresponding publishing viewpoint contents from the development data in each development day.
Confirming viewpoint types of each first-level viewpoint user, each second-level viewpoint user and each third-level viewpoint user in each developing day, wherein the viewpoint types comprise positive, neutral and negative, and simultaneously setting the weight mu of each first-level viewpoint user in each developing day tj J represents a first-order perspective user number, j=1, 2.
Counting the sum of the number of positive viewpoint users, the sum of the number of neutral viewpoint users and the sum of the number of negative viewpoint users in each developing day, respectively recorded as M Positive total t 、M Total of (F) t And M Negative total t 。
The viewpoint weights of the viewpoint users of each grade of viewpoint types positive and negative in each development day are respectively accumulated to obtain a grade of positive viewpoint weight sum mu of each development day Positive direction t Sum of first-order negative perspective weights mu Negative pole t 。
Statistics of public opinion positive trend lambda of target public opinion event in each development day Positive direction t And comparing the set positive trend degree intervals of the public opinion.
Counting the number of development days in each public opinion positive trend interval, and marking the public opinion positive trend interval with the largest number of development days as a target public opinion positive interval.
Average value calculation is carried out on the public opinion positive tendencies of all the development days in the target public opinion positive interval to obtain average public opinion positive tendenciesPublic opinion positive trend lambda for statistics of target public opinion event Positive direction ，/>ζ is a set public opinion positive trend estimation compensation factor.
Preferably, the setting of the viewpoint weight of each level of viewpoint users in each developing day includes: importing attributes of all primary viewpoint users in all developing days into attribute-level viewpoint weight assessment modelIn the method, the attribute layer view weight of each level view user in each development day is output and is recorded as +.>Take the value omega 0 Or omega 1 ，ω 0 >ω 1 。
According to the vermicelli amount of each first-class viewpoint user in each developing day, setting vermicelli-level viewpoint weight kappa of each first-class viewpoint user in each developing day tj 。
Reading quantity, praise quantity, forwarding quantity and comment user number of corresponding publishing viewpoint content of each level of viewpoint users in each developing day are respectively recorded as Y tj 、D tj 、Z tj And L tj Counting interaction layer view weight upsilon of each level view user in each developing day ij ，Y 0 、D 0 、Z 0 And L 0 Respectively setting the reading quantity, the praise quantity, the forwarding quantity and the evaluation of the referenceNumber of users is counted.
Counting viewpoint weight factor mu of each level of viewpoint users in each development day tj ，
Preferably, the specific statistical formula of the tendency of the target public opinion event on the public opinion face of each development day is:u1 and U2 are the positive trend evaluation conditions of each public opinion, and Deltaμ is the set reference opinion weight difference.
U1 represents M Total of (F) t ≥M Negative total t 、M Total of (F) t ≥M Positive total t And M Positive total t ＝M Negative total t This is true.
U2 represents M Total of (F) t ＜M Negative total t 、M Total of (F) t ＜M Positive total t And M Positive total t ≠M Negative total t This is true.
Preferably, the specific setting process of the public opinion positive trend evaluation compensation factor is as follows: the number of the development days which are more than 0 and less than 0 is respectively screened from the public opinion positive trend degree of each development day and is respectively marked as R Positive direction And R is Negative pole 。
Setting a public opinion positive trend evaluation compensation factor zeta,R positive direction -R Negative pole And < DeltaR, deltaR is the set development day deviation, and e is a natural constant.
Preferably, the setting the public opinion front trend evaluation correction factor includes: filtering out R before ranking from each development day corresponding to each similar history public opinion event 0 Each of the development days, the development days remaining after the filtration was recorded as each of the reference development days.
And extracting development data of each similar historical public opinion event corresponding to each reference development day, and carrying out homologous statistics according to the statistical mode of public opinion positive trend of the target public opinion event to obtain the public opinion positive trend of each similar historical public opinion event.
Constructing a front trend change curve of the historical similar event public opinion, positioning the number of fluctuation points from the change curve, and marking as M Wave-guide 。
And (3) overlapping and comparing the front trend change curve of the historical similar event public opinion with the front trend change curve of the reference, and counting the total length of curve segments below the front trend change curve of the reference, and marking the total length as l.
Setting a correction factor eta for public opinion positive trend evaluation,M′ wave-guide 、k Lower part(s) The number of the fluctuation points and the length ratio of the lower curve are respectively set as reference, l Total (S) The length of the change curve is trended for the public opinion of the historical similar event.
Preferably, the development tendency judgment model is specifically expressed as:p1, P2 and P3 are respectively set development tendency evaluation conditions, and P1 represents lambda' Positive direction ≥λ″ Positive direction P2 represents lambda' Positive direction ≤λ″ Positive direction P3 represents lambda Positive direction ＜λ′ Positive direction ＜λ″′ Positive direction ，λ″ Positive direction 、λ″′ Positive direction Respectively, the set tendency degree of the first public opinion face and the second public opinion face are referred to.
Compared with the prior art, the invention has the following beneficial effects: (1) According to the invention, through carrying out deep analysis on the development data of the public opinion event in each development day and combining with the development data of similar historical public opinion event, further carrying out statistics on the forward trend analysis of the public opinion event, carrying out development prediction correction of the public opinion event, judging the development trend of the public opinion event, realizing multi-aspect judgment of the public opinion event, effectively solving the defects existing in the development trend judgment of the current public opinion event, fully considering the timeliness of the public opinion event, making up the defects in the development variability analysis of the current public opinion event, facilitating public opinion related manager to grasp the opportunity of coping with the public opinion event and properly handle the potential crisis, promoting the public opinion related manager to timely adjust strategies and actions, showing the focus, the demand and the emotion change of the public, ensuring the correctness and pertinence of information provision and response of the follow-up public opinion related manager, and providing effective help for guiding the follow-up public opinion.
(2) According to the invention, through carrying out detailed analysis on the content, development data and development days of each public opinion event, similar historical event confirmation is carried out from the content and development two dimensions, and the data utilization rate of the historical related public opinion event is improved, so that the limitation and one-sided property of the current public opinion event development tendency fixed judgment basis are avoided, the reference basis of the public opinion event development tendency is expanded, the service is provided for correcting the public opinion event development tendency judgment error, and the rationality, standardization and reference property of the public opinion event development tendency judgment are ensured.
(3) According to the invention, through carrying out layer-by-layer analysis on the first-level viewpoint users, the second-level viewpoint users and the third-level viewpoint users of the public opinion event on each development day, and carrying out viewpoint weight setting on the first-level viewpoint users, the positive trend of the public opinion is further counted, the deep dynamic analysis of the public opinion event development data is realized, the defect that the current public opinion event development tendency judgment is shallow is overcome, the authenticity and reliability of the barrier public opinion event development tendency judgment result are ensured, and meanwhile, the barrier timeliness and barrier effect of the negative public opinion development are also improved.
(4) According to the method, the system and the device, the user viewpoint conditions of similar historical public opinion events after the target public opinion event development days are analyzed again, so that the public opinion positive trend evaluation correction factors are set, the error of the public opinion event development tendency judgment is reduced, the convincing and reliability of the public opinion event development tendency correction are ensured, and the target public opinion development tendency judgment is objective and accurate.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of the steps of the method of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, the present invention provides a method for determining the development tendency of a single public opinion event, which includes: step 1, extracting basic monitoring data of public opinion events: and extracting the content and the current development days of the monitored target public opinion event.
Step 2, extracting public opinion event development data: and extracting development data of the monitored target public opinion event in each development day, wherein the development data comprises opinion publishing data and interaction data.
Specifically, the viewpoint posting data includes the number of posting viewpoint users, the attribute of each posting viewpoint user, an ID account, a fan amount, and posting viewpoint content, and the interactive data includes the reading amount, the praise amount, the forwarding amount, the number of comment users, the comment content of each comment user, the number of reply users, and the reply content of each reply user corresponding to the posting viewpoint content of each posting viewpoint user.
Step 3, public opinion rating reference data analysis: and extracting the content, the number of development days and the development data in each development day of each history public opinion event from the public opinion information base, and confirming each similar history public opinion event.
Illustratively, identifying each similar historical public opinion event includes:step 3-1, respectively carrying out keyword recognition on the contents of the target public opinion event and the historical public opinion event through a keyword recognition technology to obtain keywords corresponding to the target public opinion event and the historical public opinion event, and further counting the content similarity χ of the historical public opinion event i I represents a historical public opinion event number, i=1, 2.
Understandably, the specific statistical process for counting the content similarity corresponding to each public opinion event is as follows: and forming a keyword set of the target public opinion event by each keyword of the target public opinion event, and marking the keyword set as A.
B, recording the keyword of each public opinion event to form a keyword set of each public opinion event i Further, content similarity χ corresponding to each public opinion event of the history is calculated i ，
In one embodiment, the keyword recognition technology is a mature technology, and further description is omitted.
Step 3-2, the current development day number of the target public opinion event is recorded as R 0 。
Step 3-3, sequencing the development days corresponding to the historical public opinion events according to time sequence, and further extracting R before ranking from the development data in the development days corresponding to the historical public opinion events 0 Development data of each development day, and statistics of development similarity beta of each public opinion event i 。
Further, the statistics of the development similarity of each public opinion event includes: step 3-3-1, recording the number of the development days of each public opinion event as R' i 。
Step 3-3-2, if R' i ≤R 0 The i-th historical public opinion event is marked as a non-similar public opinion event, and the development similarity of the non-similar public opinion event is marked as
Step 3-3-3, if R' i ＞R 0 And the ith historical public opinion event is recorded as a reference public opinion event.
Step 3-3-4, numbering each development day of the target public opinion event, and extracting the number M of users publishing opinion from the development data in each development day 0 t T represents a day of development number, t=1, 2.
Step 3-3-5, ranking the reference public opinion event corresponding to R 0 The development data of each development day is used as reference development data of target public opinion event corresponding to each development day, and the number M of users publishing opinion is extracted from the reference development data 1 t 。
Step 3-3-6, extracting attributes of users from each publishing viewpoint from development data in each development day, and counting the user distribution richness F of each development day t Mass ratio k Group of t And a media duty k Media (media) t 。
It should be noted that, the specific statistical process of the user distribution richness, the crowd occupation ratio and the media occupation ratio of each developing day is as follows: and comparing the attributes of the publishing viewpoint users in each developing day to obtain the attribute type number of the publishing viewpoint users and the attribute type publishing viewpoint users.
Comparing the number of the attribute types of the users of the published views in each developing day with the number of the attribute types of the users set by the social platform to obtain the distribution richness of the users in each developing day.
And locating the number of the publishing viewpoint users with the attribute types of masses from the number of the publishing viewpoint users with the attribute types of each attribute type in each development day, and comparing the number of the publishing viewpoint users with the target public opinion event in each development day to obtain the mass ratio of each development day.
And locating the number of the publishing viewpoint users with the attribute type as media from the number of the publishing viewpoint users with the attribute type in each development day, and comparing the number of the publishing viewpoint users with the target public opinion event in each development day to obtain the media duty ratio of each development day.
Step 3-3-7, according to F, according to the reference development data in each development day t 、k Group of t 、k Media (media) t Statistics of user distribution richness F of reference in each development day t ' ratio of massesAnd a media duty k' Media (media) 。
Step 3-3-8, counting the similarity XY of the participating users referring to the public opinion,，ΔM、ΔF、Δk group of 、Δk Media (media) The number of participating users with the permission of setting similar public opinion event is poor, the distribution richness of users is deviated, the mass ratio is deviated, the media ratio is deviated, +.>Representing rounding down symbols.
Extracting the reading quantity, the praise quantity, the forwarding quantity and the comment user number of the corresponding publishing viewpoint content of each publishing viewpoint user from the development data in each development day, and counting the development heat tendency degree in each development day, wherein the trend degree is marked as phi t And similarly counting the trend phi 'of the reference development heat in each development day' t 。
It should be noted that, the specific statistical process of the trend of the development heat in each development day is as follows: respectively carrying out average calculation on the reading quantity, the praise quantity, the forwarding quantity and the comment user number of the corresponding publishing viewpoint content of each publishing viewpoint user in each developing day to obtain the average comprehensive reading quantity, the average praise quantity, the average comprehensive forwarding quantity and the average comprehensive comment user number corresponding to the single publishing viewpoint user in each developing day, and respectively recording the average comprehensive reading quantity, the average praise quantity, the average comprehensive forwarding quantity and the average comprehensive comment user number as followsAnd->
Calculating trend phi of development heat in each development day t ，Y ', D', Z 'and L' are the average reading amount, the average praise amount, the average forwarding amount and the average comment user number of the set reference respectively.
Step 3-3-9, counting the development heat similarity FD of the reference public opinion event, further counting the development similarity psi of the reference public opinion event,XY ', FD' are respectively set up a reference participating user similarity threshold and a development heat similarity threshold, so as to obtain the development similarity beta of each public opinion event of history i ，β i The value is +.>Or psi, & gt>
It should be noted that the number of the substrates,delta phi is the deviation of the trend of the set development heat.
Step 3-4, counting the comprehensive public opinion similarity gamma corresponding to each public opinion event of the history i ，χ 'and β' are respectively the content similarity and the development similarity of the set reference, a1 and a2 are respectively the content similarity and the development similarity corresponding to the comprehensive public opinion similarity evaluation duty ratio weight, and σ is the set public opinion similarity evaluation correction factor.
Step 3-5, gamma is added i And comparing the evaluation similarity gamma 'of the set similar public opinion events, and screening out each historical public opinion event larger than gamma' as each similar historical public opinion event.
According to the embodiment of the invention, through carrying out detailed analysis on the content, development data and development days of each public opinion event, similar historical event confirmation is carried out from the content and development two dimensions, and the data utilization rate of the historical related public opinion event is improved, so that the limitation and one-sided property of the current public opinion event development tendency fixed judgment basis are avoided, the reference basis of the public opinion event development tendency is expanded, the service is provided for correcting the public opinion event development tendency judgment error, and the rationality, normalization and reference property of the public opinion event development tendency judgment are ensured.
And 4, public opinion event development prediction analysis: public opinion positive trend lambda for analyzing target public opinion event Positive direction 。
Illustratively, analyzing the public opinion positive trend of the target public opinion event includes: and 4-1, respectively marking the publishing viewpoint user, the comment user and the reply user as a first viewpoint user, a second viewpoint user and a third viewpoint user.
And 4-2, extracting comment contents of the secondary viewpoint users and reply contents of the tertiary viewpoint users under the corresponding publishing viewpoint contents from the development data in each development day.
Step 4-3, confirming viewpoint types of each first-level viewpoint user, each second-level viewpoint user and each third-level viewpoint user in each developing day, wherein the viewpoint types comprise positive, neutral and negative, and simultaneously setting the weight mu of each first-level viewpoint user in each developing day tj J represents a first-order perspective user number, j=1, 2.
The identifying of the viewpoint types of each primary viewpoint user, each secondary viewpoint user, and each tertiary viewpoint user in each developing day includes: and E1, extracting keywords from the published viewpoint contents corresponding to the first-level viewpoint users in each developing day through a keyword extraction technology, and obtaining viewpoint keywords of the published viewpoint contents corresponding to the first-level viewpoint users in each developing day.
And E2, matching and comparing the viewpoint keywords corresponding to the content of the published viewpoints of the first-level viewpoint users in each developing day with a set viewpoint keyword set corresponding to each viewpoint type, and counting the number of the viewpoint keywords corresponding to each viewpoint type of the first-level viewpoint users in each developing day.
E3, screening the viewpoint keyword number C with the viewpoint type being positive from the viewpoint keyword number corresponding to the viewpoint type of each first-class viewpoint user in each developing day tj View key number C 'with view type being neutral' tj And the number of viewpoint keywords C' for which the viewpoint type is negative tj 。
E4, ifAnd judging the view type corresponding to the jth level view user in the t developing day as the front.
E5, ifOr C' tj ＞C′ tj And C' tj ＞C tj Or C tj ＝C′ tj ＝C″ tj If so, judging that the view type corresponding to the jth level view user in the t developing day is neutral.
E6, ifAnd judging that the view type corresponding to the jth level of view user in the t developing days is negative, thereby obtaining the view type of each level of view user in each developing day.
And E7, analyzing the viewpoint types of the secondary viewpoint users and the tertiary viewpoint users in a same way according to the analysis mode of the viewpoint types of the primary viewpoint users in each developing day.
Further, setting the viewpoint weight of each level of viewpoint users in each developing day includes: k1, importing the attribute of each level of view user in each developing day into an attribute-level view weight evaluation modelIn the method, the attribute layer view weight of each level view user in each developing day is output and is recorded as/>Take the value of omega 0 or omega 1 ，ω 0 >ω 1 。
In a specific embodiment ω 0 The value can be 0.8 omega 1 The value can be 0.5.
K2, setting vermicelli layer viewpoint weights kappa of the first-level viewpoint users in each developing day according to the vermicelli amounts Stj of the first-level viewpoint users in each developing day tj ，S 0 The reference vermicelli amount is set.
K3, respectively marking the reading quantity, the praise quantity, the forwarding quantity and the comment user number of the corresponding published viewpoint content of each level viewpoint user in each development day as Y tj 、D tj 、Z tj And L tj Counting the interaction layer view weight uij of each level view user in each developing day,Y 0 、D 0 、Z 0 and L 0 The reading quantity, the praise quantity, the forwarding quantity and the comment user number of the setting reference are respectively.
K4, counting viewpoint weight factors mu of each level of viewpoint users in each development day tj ，Is indicated at->υ ij 、κ tj And takes the maximum value.
Step 4-4, counting the sum of the number of positive viewpoint users, the sum of the number of neutral viewpoint users and the sum of the number of negative viewpoint users in each developing day, respectively marked as M Positive total t 、M Total of (F) t And M Negative total t 。
It should be noted that, the statistical process of the total number of positive viewpoint users, the total number of neutral viewpoint users and the total number of negative viewpoint users in each developing day is as follows: w1, respectively taking viewpoint type as a primary viewpoint user, a secondary viewpoint user and a tertiary viewpoint user, counting the number of the primary viewpoint user, the secondary viewpoint user and the tertiary viewpoint user in each development day, and respectively marking as M Positive direction t 、And->
W2, counting sum M of number of users in front views in each developing day Positive total t ，b1, b2 and b3 are respectively set primary viewpoint users, secondary viewpoint users and tertiary viewpoint users and correspond to the number evaluation duty ratio weight.
In one embodiment, b1, b2, b3 may take values of 0.5, 0.3, 0.2, respectively.
W3, counting the number of the one-stage neutral viewpoint user, the two-stage neutral viewpoint user and the three-stage neutral viewpoint user in each development day by taking the viewpoint type as the one-stage neutral viewpoint user, the two-stage neutral viewpoint user and the three-stage neutral viewpoint user, and recording the number as M respectively In (a) t 、And->
W4, counting sum M of neutral viewpoint users in each developing day Total of (F) t ，
W5, respectively taking viewpoint type as a first-level negative viewpoint user, a second-level negative viewpoint user and a third-level negative viewpoint user, counting the number of the first-level negative viewpoint user, the second-level negative viewpoint user and the third-level negative viewpoint user in each development day, and respectively marking as M Negative pole t 、And->
W6, counting the sum M of the negative viewpoint users in each developing day Negative total t ，
Step 4-5, accumulating the viewpoint weights of the viewpoint users of each level with positive and negative viewpoint types in each development day respectively to obtain the sum mu of the viewpoint weights of the first level positive in each development day Positive direction t Sum of first-order negative perspective weights mu Negative pole t 。
Step 4-6, counting public opinion positive trend lambda of target public opinion event in each development day Positive direction t And comparing the set positive trend degree intervals of the public opinion.
Further, a specific statistical formula of the public opinion positive trend of the target public opinion event in each development day is as follows:u1 and U2 are the positive trend evaluation conditions of each public opinion, and Deltaμ is the set reference opinion weight difference.
U1 represents M Total of (F) t ≥M Negative total t 、M Total of (F) t ≥M Positive total t And M Positive total t ＝M Negative total t This is true.
U2 represents M Total of (F) t ＜M Negative total t 、M Total of (F) t ＜M Positive total t And M Positive total t ≠M Negative total t This is true.
And 4-7, counting the number of development days in each public opinion positive trend interval, and marking the public opinion positive trend interval with the largest number of development days as a target public opinion positive interval.
Step 4-8, carrying out average calculation on the public opinion positive tendencies of all the development days in the target public opinion positive interval to obtain average public opinion positive tendenciesPublic opinion positive trend lambda for statistics of target public opinion event Positive direction ，/>ζ is a set public opinion positive trend estimation compensation factor.
Specifically, the specific setting process of the public opinion positive trend evaluation compensation factor is as follows: the number of the development days which are more than 0 and less than 0 is respectively screened from the public opinion positive trend degree of each development day and is respectively marked as R Positive direction And R is Negative pole 。
Setting a public opinion positive trend evaluation compensation factor zeta,R positive direction -R Negative pole And < DeltaR, deltaR is the set development day deviation, and e is a natural constant.
According to the embodiment of the invention, the first-level viewpoint users, the second-level viewpoint users and the third-level viewpoint users of the public opinion event in each development day are analyzed layer by layer, viewpoint weights of the first-level viewpoint users are set, and then the positive trend of the public opinion is counted, so that the deep dynamic analysis of the public opinion event development data is realized, the defect that the current public opinion event development tendency judgment is shallow is overcome, the authenticity and the reliability of the barrier public opinion event development tendency judgment result are ensured, and meanwhile, the barrier timeliness and the barrier effect of the negative public opinion development are also improved.
Step 5, public opinion event development prediction correction: setting a public opinion positive trend evaluation correction factor eta so as to predict the estimated public opinion positive trend lambda 'of the target public opinion event' Positive direction ，λ′ Positive direction ＝λ Positive direction *(1-η)。
Specifically, setting a public opinion positive trend evaluation correction factor includes: step 5-1, filtering out R before ranking from the corresponding developing days of the similar historical public opinion events 0 Each of the development days, the development days remaining after the filtration was recorded as each of the reference development days.
And 5-2, extracting development data of each similar historical public opinion event corresponding to each reference development day, and carrying out homography statistics according to the statistical mode of public opinion positive tendencies of the target public opinion event to obtain the public opinion positive tendencies of each similar historical public opinion event.
Step 5-3, constructing a change curve of the public opinion positive trend of the historical similar event by taking the similar historical public opinion event as a horizontal axis and the public opinion positive trend degree as a vertical axis, positioning the number of fluctuation points from the change curve, and marking as M Wave-guide 。
In one embodiment, a fluctuation point refers to a point on a curve that is opposite in left-right direction, such as a point that rises on the left, falls on the right, or rises on the left, falls on the right.
And 5-4, overlapping and comparing the front trend change curve of the history similar event public opinion with the front trend change curve of the reference, and counting the total length of the curve segment below the front trend change curve of the reference, and marking the total length as l.
Setting a correction factor eta for public opinion positive trend evaluation,M′ wave-guide 、k Lower part(s) The number of the fluctuation points and the length ratio of the lower curve are respectively set as reference, l Total (S) The length of the change curve is trended for the public opinion of the historical similar event.
According to the embodiment of the invention, through re-analyzing the user viewpoint situation of similar historical public opinion events after the target public opinion event development days, the public opinion positive trend evaluation correction factor is set, and the error of the public opinion event development tendency judgment is reduced, so that the convincing and reliability of the public opinion event development tendency correction are ensured, and the target public opinion development tendency judgment is objective and accurate.
Step 6, public opinion event development tendency judgment: will be lambda' Positive direction And importing the target public opinion event into a development tendency judgment model, and outputting the development tendency of the target public opinion event.
Specifically, the development tendency judgment model is specifically expressed as:p1, P2 and P3 are respectively set development tendency evaluation conditions, and P1 represents lambda' Positive direction ≥λ″ Positive direction P2 represents lambda' Positive direction ≤λ″ Positive direction P3 represents lambda Positive direction ＜λ′ Positive direction ＜λ″ Positive direction ，λ″ Positive direction 、λ″′ Positive direction Respectively, the set tendency degree of the first public opinion face and the second public opinion face are referred to.
According to the embodiment of the invention, through carrying out deep analysis on the development data of the public opinion event in each development day and combining with the development data of similar historical public opinion events, further carrying out statistics on the forward trend analysis of the public opinion event, carrying out development prediction correction of the public opinion event, judging the development trend of the public opinion event, realizing multi-aspect judgment of the public opinion event, effectively solving the defect of judgment of the development trend of the current public opinion event, fully considering the timeliness of the public opinion event, making up the defect in the development variability analysis of the current public opinion event, facilitating public opinion related management personnel to grasp the opportunity of coping with the public opinion event and properly handle the potential crisis, promoting the public opinion related management personnel to timely adjust strategies and actions, showing the focus, the demand and the emotion change of the public, ensuring the correctness and pertinence of information provision and response of the follow-up public opinion related management personnel, and providing effective help for the guidance of the follow-up public opinion guidance.
Step 7, public opinion event development tendency feedback: and feeding back the development tendency of the target public opinion event to relevant management and control personnel of the target public opinion event.
The foregoing is merely illustrative and explanatory of the principles of this invention, as various modifications and additions may be made to the specific embodiments described, or similar arrangements may be substituted by those skilled in the art, without departing from the principles of this invention or beyond the scope of this invention as defined in the claims.
Claims (10)
1. A single public opinion event development tendency judging method is characterized in that: comprising the following steps:
step 1, extracting basic monitoring data of public opinion events: extracting the content and the current development days of the monitored target public opinion event;
step 2, extracting public opinion event development data: extracting development data of the monitored target public opinion event in each development day;
step 3, public opinion rating reference data analysis: extracting the content, the number of development days and the development data in each development day of each history public opinion event from the public opinion information base, and confirming each similar history public opinion event;
and 4, public opinion event development prediction analysis: public opinion positive trend lambda for analyzing target public opinion event Positive direction ；
Step 5, public opinion event development prediction correction: setting a public opinion positive trend evaluation correction factor eta so as to predict the estimated public opinion positive trend lambda 'of the target public opinion event' Positive direction ，λ′ Positive direction ＝λ Positive direction *(1-η)；
Step 6, public opinion event development tendency judgment: will be lambda' Positive direction Leading in a development tendency judgment model, and outputting the development tendency of the target public opinion event;
step 7, public opinion event development tendency feedback: and feeding back the development tendency of the target public opinion event to relevant management and control personnel of the target public opinion event.
2. The method of claim 1, wherein the single public opinion event development tendency determination is characterized by: the development data comprise view publishing data and interaction data;
the viewpoint posting data comprises the number of posting viewpoint users, the attribute of each posting viewpoint user, an ID account number, a fan amount and posting viewpoint contents, and the interactive data comprises the reading amount, the praise amount, the forwarding amount, the number of comment users, the comment contents of each comment user, the number of reply users and the reply contents of each reply user corresponding to the posting viewpoint contents of each posting viewpoint user.
3. The method of claim 2, wherein the single public opinion event development tendency determination is characterized by: the identifying each similar historical public opinion event includes:
keyword recognition is respectively carried out on the contents of the target public opinion event and the historical public opinion event through keyword recognition technology, so that keywords corresponding to the target public opinion event and the historical public opinion event are obtained, and then the content similarity χ of the historical public opinion event is counted i I represents a historical public opinion event number, i=1, 2,..;
the current development day number of the target public opinion event is recorded as R 0 ；
Sequencing the development days corresponding to the historical public opinion events according to time sequence, and extracting R before ranking from the development data in the development days corresponding to the historical public opinion events 0 Development data of each development day, and statistics of development similarity beta of each public opinion event i ；
Statistics of comprehensive public opinion similarity gamma corresponding to each public opinion event of history i ，χ 'and β' are respectively the content similarity and the development similarity of the set reference, a1 and a2 are respectively the content similarity and the development similarity corresponding to the comprehensive public opinion similarity evaluation duty ratio weight, and σ is the set public opinion similarity evaluation correction factor>Representing a downward rounding symbol;
will gamma i And comparing the evaluation similarity gamma 'of the set similar public opinion events, and screening out each historical public opinion event larger than gamma' as each similar historical public opinion event.
4. The method for determining the propensity of a single public opinion event to develop according to claim 3, wherein: the statistics of the development similarity of each public opinion event includes:
the number of the development days of each public opinion event is recorded as R' i ；
If R 'is' i ≤R 0 The i-th historical public opinion event is marked as a non-similar public opinion event, and the development similarity of the non-similar public opinion event is marked as
If R 'is' i ＞R 0 The i-th historical public opinion event is recorded as a reference public opinion event;
numbering each development day of the target public opinion event, and extracting the number M of users publishing opinion from the development data in each development day 0 t T represents a day of development number, t=1, 2,..p;
will refer to public opinion event corresponding top R 0 The development data of each development day is used as reference development data of target public opinion event corresponding to each development day, and the number M of users publishing opinion is extracted from the reference development data 1 t ；
Extracting attributes of users from the published views from the development data in each development day, and counting the user distribution richness F of each development day t Mass ratio k Group of t And a media duty k Media (media) t ；
According to reference development data in each development day, according to F t 、k Group of t 、k Media (media) t Statistics of user distribution richness F of reference in each development day t ' and massesDuty ratio k' Group of t And a media duty k' Media (media) ；
The similarity XY of participating users with reference to public opinion is counted,，ΔM、ΔF、Δk group of 、Δk Media (media) The method comprises the steps of respectively setting the difference of the number of participating users permitted by similar public opinion events, the deviation of the abundance of user distribution, the deviation of the mass ratio and the deviation of the media ratio;
extracting the reading quantity, the praise quantity, the forwarding quantity and the comment user number of the corresponding publishing viewpoint content of each publishing viewpoint user from the development data in each development day, and counting the development heat tendency degree in each development day, wherein the trend degree is marked as phi t And similarly counting the trend phi 'of the reference development heat in each development day' t ；
The development heat similarity FD of the reference public opinion event is counted, and then the development similarity psi of the reference public opinion event is counted,XY ', FD' are respectively set up a reference participating user similarity threshold and a development heat similarity threshold, so as to obtain the development similarity beta of each public opinion event of history i ，β i The value is +.>Or a combination of the above-mentioned materials,
5. the method of claim 2, wherein the single public opinion event development tendency determination is characterized by: the analyzing the public opinion positive trend degree of the target public opinion event comprises the following steps:
marking the publishing viewpoint user, the comment user and the reply user as a first-level viewpoint user, a second-level viewpoint user and a third-level viewpoint user respectively;
extracting comment contents of each secondary viewpoint user and reply contents of each tertiary viewpoint user under corresponding publishing viewpoint contents from the development data in each development day;
confirming viewpoint types of each first-level viewpoint user, each second-level viewpoint user and each third-level viewpoint user in each developing day, wherein the viewpoint types comprise positive, neutral and negative, and simultaneously setting the weight mu of each first-level viewpoint user in each developing day tj J represents a first-order perspective user number, j=1, 2.
Counting the sum of the number of positive viewpoint users, the sum of the number of neutral viewpoint users and the sum of the number of negative viewpoint users in each developing day, respectively recorded as M Positive total t 、M Total of (F) t And M Negative total t ；
The viewpoint weights of the viewpoint users of each grade of viewpoint types positive and negative in each development day are respectively accumulated to obtain a grade of positive viewpoint weight sum mu of each development day Positive direction t Sum of first-order negative perspective weights mu Negative pole t ；
Statistics of public opinion positive trend lambda of target public opinion event in each development day Positive direction t Comparing the set positive trend degree intervals of the public opinion with the set positive trend degree intervals of the public opinion;
counting the number of development days in each public opinion positive trend interval, and marking the public opinion positive trend interval with the largest number of development days as a target public opinion positive interval;
average value calculation is carried out on the public opinion positive tendencies of all the development days in the target public opinion positive interval to obtain average public opinion positive tendenciesPublic opinion positive trend lambda for statistics of target public opinion event Positive direction ，/>ζ is a set public opinion positive trend estimation compensation factor.
6. The method of claim 5, wherein the single public opinion event development tendency determination is characterized by: the setting of the viewpoint weight of each level of viewpoint users in each developing day comprises the following steps:
importing attributes of all primary viewpoint users in all developing days into attribute-level viewpoint weight assessment modelIn the method, the attribute layer view weight of each level view user in each development day is output and is recorded as +.>Take the value of omega 0 or omega 1 ，ω 0 >ω 1 ；
According to the vermicelli amount of each first-class viewpoint user in each developing day, setting vermicelli-level viewpoint weight kappa of each first-class viewpoint user in each developing day tj ；
Reading quantity, praise quantity, forwarding quantity and comment user number of corresponding publishing viewpoint content of each level of viewpoint users in each developing day are respectively recorded as Y tj 、D tj 、Z tj And L tj Counting the interaction layer view weight uij of each level view user in each developing day,Y 0 、D 0 、Z 0 and L 0 Respectively setting a reference reading quantity, a praise quantity, a forwarding quantity and the number of comment users;
counting viewpoint weight factors mu tj of each level of viewpoint users in each developing day,
7. the method of claim 5, wherein the single public opinion event development tendency determination is characterized by: the specific statistical formula of the public opinion positive trend degree of the target public opinion event in each development day is as follows:
u2, U1 and U2 are the trend evaluation conditions of the front of each public opinion, and Deltamu is the set reference viewpoint weight difference;
u1 representsM Total of (F) t ≥M Positive total t And M Positive total t ＝M Negative total t Establishment;
u2 representsM Total of (F) t ＜M Positive total t And M Positive total t ≠M Negative total t This is true.
8. The method of claim 5, wherein the single public opinion event development tendency determination is characterized by: the specific setting process of the public opinion positive trend evaluation compensation factor comprises the following steps:
the number of the development days which are more than 0 and less than 0 is respectively screened from the public opinion positive trend degree of each development day and is respectively marked as R Positive direction And R is Negative pole ；
Setting a public opinion positive trend evaluation compensation factor zeta,R positive direction -R Negative pole And < DeltaR, deltaR is the set development day deviation, and e is a natural constant.
9. The method of claim 6, wherein the single public opinion event development tendency determination is characterized by: the setting of the public opinion positive trend evaluation correction factor comprises the following steps:
from various similar historic public opinion mattersFiltering out R before ranking in each corresponding developing day 0 Each development day is marked as each reference development day;
extracting development data of each similar historical public opinion event corresponding to each reference development day, and carrying out homography statistics according to the statistical mode of public opinion positive trend of the target public opinion event to obtain the public opinion positive trend of each similar historical public opinion event;
constructing a front trend change curve of the historical similar event public opinion, positioning the number of fluctuation points from the change curve, and marking as M Wave-guide ；
The front trend change curve of the historical similar event public opinion is overlapped with the reference front trend change curve, the total length of the curve segment below the reference front trend change curve is counted and recorded as l Lower part(s) ；
Setting a correction factor eta for public opinion positive trend evaluation,M′ wave-guide 、k Lower part(s) The number of the fluctuation points and the length ratio of the lower curve are respectively set as reference, l Total (S) The length of the change curve is trended for the public opinion of the historical similar event.
10. The method of claim 1, wherein the single public opinion event development tendency determination is characterized by: the development tendency judgment model is specifically expressed as follows:p1, P2 and P3 are respectively set development tendency evaluation conditions, and P1 represents lambda' Positive direction ≥λ″′ Positive direction P2 represents lambda' Positive direction ≤λ″ Positive direction P3 represents lambda Positive direction ＜λ′ Positive direction ＜λ″′ Positive direction ，λ″ Positive direction 、λ″′ Positive direction Respectively, the set tendency degree of the first public opinion face and the second public opinion face are referred to.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311512337.6A CN117494897B (en) | 2023-11-14 | Single public opinion event development tendency judging method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311512337.6A CN117494897B (en) | 2023-11-14 | Single public opinion event development tendency judging method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117494897A true CN117494897A (en) | 2024-02-02 |
| CN117494897B CN117494897B (en) | 2024-05-17 |
Family
ID=
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106156257A (en) * | 2015-04-28 | 2016-11-23 | 北大方正集团有限公司 | A kind of Tendency Prediction method of microblogging public sentiment event |
| WO2019227710A1 (en) * | 2018-05-31 | 2019-12-05 | 平安科技（深圳）有限公司 | Network public opinion analysis method and apparatus, and computer-readable storage medium |
| CN111026868A (en) * | 2019-12-05 | 2020-04-17 | 厦门市美亚柏科信息股份有限公司 | Multi-dimensional public opinion crisis prediction method, terminal device and storage medium |
| CN111428113A (en) * | 2020-03-27 | 2020-07-17 | 华侨大学 | Network public opinion guiding effect prediction method based on fuzzy comprehensive evaluation |
| WO2021073271A1 (en) * | 2019-10-17 | 2021-04-22 | 平安科技（深圳）有限公司 | Public opinion analysis method and device, computer device and storage medium |
| CN114385912A (en) * | 2021-12-25 | 2022-04-22 | 西安康奈网络科技有限公司 | Method for judging place where internet public opinion information occurs |
| CN115269950A (en) * | 2022-06-07 | 2022-11-01 | 青岛理工大学 | Public opinion information content mining and propagation monitoring analysis method |
| CN115658996A (en) * | 2022-11-11 | 2023-01-31 | 苏州华必讯信息科技有限公司 | Public opinion monitoring method with emergency event analysis and extraction function |
| WO2023029462A1 (en) * | 2021-08-31 | 2023-03-09 | 西南电子技术研究所（中国电子科技集团公司第十研究所） | Hot event state evaluation method |
| CN116992146A (en) * | 2023-08-10 | 2023-11-03 | 北京橙色风暴数字技术有限公司 | Public opinion monitoring system and method based on big data |
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106156257A (en) * | 2015-04-28 | 2016-11-23 | 北大方正集团有限公司 | A kind of Tendency Prediction method of microblogging public sentiment event |
| WO2019227710A1 (en) * | 2018-05-31 | 2019-12-05 | 平安科技（深圳）有限公司 | Network public opinion analysis method and apparatus, and computer-readable storage medium |
| WO2021073271A1 (en) * | 2019-10-17 | 2021-04-22 | 平安科技（深圳）有限公司 | Public opinion analysis method and device, computer device and storage medium |
| CN111026868A (en) * | 2019-12-05 | 2020-04-17 | 厦门市美亚柏科信息股份有限公司 | Multi-dimensional public opinion crisis prediction method, terminal device and storage medium |
| CN111428113A (en) * | 2020-03-27 | 2020-07-17 | 华侨大学 | Network public opinion guiding effect prediction method based on fuzzy comprehensive evaluation |
| WO2023029462A1 (en) * | 2021-08-31 | 2023-03-09 | 西南电子技术研究所（中国电子科技集团公司第十研究所） | Hot event state evaluation method |
| CN114385912A (en) * | 2021-12-25 | 2022-04-22 | 西安康奈网络科技有限公司 | Method for judging place where internet public opinion information occurs |
| CN115269950A (en) * | 2022-06-07 | 2022-11-01 | 青岛理工大学 | Public opinion information content mining and propagation monitoring analysis method |
| CN115658996A (en) * | 2022-11-11 | 2023-01-31 | 苏州华必讯信息科技有限公司 | Public opinion monitoring method with emergency event analysis and extraction function |
| CN116992146A (en) * | 2023-08-10 | 2023-11-03 | 北京橙色风暴数字技术有限公司 | Public opinion monitoring system and method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Goletsis et al. | Project ranking in the Armenian energy sector using a multicriteria method for groups | |
| CN106779457A (en) | A kind of rating business credit method and system | |
| Redman | Measuring data accuracy: A framework and review | |
| CN110008254B (en) | Transformer equipment standing book checking processing method | |
| CN111178623B (en) | Business process remaining time prediction method based on multilayer machine learning | |
| CN112232909A (en) | Business opportunity mining method based on enterprise portrait | |
| CN108596678A (en) | A kind of airline passenger value calculation method | |
| CN109934469A (en) | Based on the heterologous power failure susceptibility method for early warning and device for intersecting regression analysis | |
| CN116011827A (en) | Power failure monitoring analysis and early warning system and method for key cells | |
| CN116029684A (en) | Analysis method and system for measuring matching degree of workers | |
| CN117494897B (en) | Single public opinion event development tendency judging method | |
| CN117494897A (en) | Single public opinion event development tendency judging method | |
| Mokhtar et al. | Combination of AHP-PROMETHEE and TOPSIS for selecting the best Demand Side Management (DSM) options | |
| CN114219245B (en) | Rural power index evaluation method and device based on big data and storage medium | |
| Lee et al. | Towards discovering emerging technologies based on decision tree | |
| Bradley et al. | The philosophy of climate science | |
| CN114493364A (en) | Model construction method and device, computer readable storage medium and electronic equipment | |
| CN113506190A (en) | Abnormal electricity consumption behavior identification method, device, equipment and storage medium | |
| CN114385403A (en) | Distributed cooperative fault diagnosis method based on double-layer knowledge graph framework | |
| CN113379211A (en) | Block chain-based logistics information platform default risk management and control system and method | |
| Herrmann et al. | The statistical value chain-a benchmarking checklist for decision makers to evaluate decision support seen from a statistical point-of-view | |
| CN116029687A (en) | Intelligent talent selection evaluation analysis management system in enterprise | |
| CN114266503A (en) | Data quality evaluation method for infrastructure ERP system | |
| CN112488572B (en) | Audit object recommendation method, device, equipment and medium | |
| CN112215494A (en) | House lease identification method and system for residential electricity |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |