CN109472415A - A method of event scale in social media is predicted by kinetic characteristics - Google Patents
A method of event scale in social media is predicted by kinetic characteristics Download PDFInfo
- Publication number
- CN109472415A CN109472415A CN201811359804.5A CN201811359804A CN109472415A CN 109472415 A CN109472415 A CN 109472415A CN 201811359804 A CN201811359804 A CN 201811359804A CN 109472415 A CN109472415 A CN 109472415A
- Authority
- CN
- China
- Prior art keywords
- event
- temperature
- sampling
- social media
- basic
- 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
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000005070 sampling Methods 0.000 claims abstract description 41
- 230000033001 locomotion Effects 0.000 claims abstract description 15
- 239000000284 extract Substances 0.000 claims abstract description 5
- 230000001133 acceleration Effects 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000011161 development Methods 0.000 abstract description 23
- 230000018109 developmental process Effects 0.000 description 21
- 238000009792 diffusion process Methods 0.000 description 8
- 241000282326 Felis catus Species 0.000 description 5
- 210000000689 upper leg Anatomy 0.000 description 5
- 238000000605 extraction Methods 0.000 description 3
- 230000003211 malignant effect Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 241001425718 Vagrans egista Species 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 241000251468 Actinopterygii Species 0.000 description 1
- 241000700159 Rattus Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 235000021190 leftovers Nutrition 0.000 description 1
- 201000005630 leukorrhea Diseases 0.000 description 1
- 235000012054 meals Nutrition 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000008929 regeneration Effects 0.000 description 1
- 238000011069 regeneration method Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 206010046901 vaginal discharge Diseases 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
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of 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)
- Investigating Or Analyzing Materials Using Thermal Means (AREA)
Abstract
The invention discloses a kind of methods for predicting event scale in social media by kinetic characteristics, the following steps are included: (1) carries out unsupervised merging to social media content, generate event set, the temperature of each event in event set is calculated again, finally extracts the basic temperature event for meeting basic temperature thresholding；(2) variable cycle sampling is carried out to the temperature of basic temperature event, accelerator or moderating process are in by the temperature that variable motion rule calculates basic temperature event after sampling, if the temperature of basic temperature event is in accelerator, whether there is the possibility as mass incident by the prediction of result of the variable motion basis temperature event and be likely to become the arrival time of mass incident.Event development is mapped as variable motion by the present invention, is to be expected with the possible development scale of event, calculability and arrival time, more directly effective measuring and calculating in real time can be carried out to event scale, more efficient, precision is more quasi-.
Description
Technical field
The present invention relates to information technology fields, more particularly to one kind to pass through kinetic characteristics and predict that event is advised in social media
The method of mould.
Background technique
It is epoch from media as the quick high speed development of mobile Internet enters everybody, interaction is more real-time just
Victory, cost are cheaper.No matter the big mishap on network or in reality is by the quick wide-scale distribution of social media platform,
Social life, policies and regulations, the influence of people's phychology are gradually deepened.The problems such as event-monitoring, Public-opinion directing, causes government at different levels
The great attention of part and enterprises and institutions.As a result, to the real-time monitoring of social media, especially not to " potential risk " event
To develop, to carry out prediction be event-monitoring, mostly important reference in Public-opinion directing work.
Currently, existing event scale forecast is mainly expanded from event based on summary, the classification after a large amount of events occur
The mode of dissipating, the angle of the mode of intelligence transmission form experience and model, and are used for event scale forecast.Wherein, it is spread in event
The anxious narrow hairdo diffusion, diffusion of depth sprawling formula, the diffusion of region displacement-type, the diffusion of heterogeneous conversion type, chain is proposed in mode
Formula diffusion, circulating diffusion, radiant type diffusion etc.；Single-stranded transmitting, tree-shaped transmitting, netted biography are proposed in the mode of intelligence transmission
It passs.
But with the development speed that internet makes rapid progress, there are new type, new rule event diffusion and propagate can
Energy.At this point, according to the traditional approach summarized again is first accumulated, because it needs the more time to accumulate for emerging event so as to correlation point
Analysis and summarize, then will necessarily exist a lag period, such vacuum space every, it is most likely that influence event development is sentenced
It is disconnected, the best opportunity of Public-opinion directing is missed, event nature malignant development is caused, influences the normal stable order of society.
It is therefore desirable to which one kind does not depend on supervised learning process, no lag period can highly be adapted to over, present, Yi Jiwei
Carry out the new method of all kinds of event rules of development.
Summary of the invention
It is an object of the invention to solve the above-mentioned problems in the prior art, provide a kind of pre- by kinetic characteristics
Event development is mapped as variable motion by the method for surveying event scale in social media, the present invention, with the possible development rule of event
Mould is to be expected, calculability and arrival time, more directly effective measuring and calculating in real time can be carried out to event scale, efficiency is more
Height, precision are more quasi-.
To achieve the above object, The technical solution adopted by the invention is as follows:
A method of event scale in social media is predicted by kinetic characteristics, it is characterised in that including following step
It is rapid:
(1) unsupervised merging is carried out to social media content, generates event set, then calculate the heat of each event in event set
Degree finally extracts the basic temperature event for meeting basic temperature thresholding；Wherein, the temperature of event by reading number, comment number, turn
Hair number weighted comprehensive is calculated；
(2) variable cycle sampling is carried out to the temperature of basic temperature event, is calculated after sampling by variable motion rule
The temperature of basic temperature event is in accelerator or moderating process, if the temperature of basic temperature event is in accelerator,
Whether there is the possibility as mass incident by the prediction of result of the variable motion basis temperature event and is likely to become
The arrival time of mass incident.
Unsupervised merging in the step (1) refers to: first segmenting to social media content, and passes through TF-IDF
Feature Words extraction is carried out, when occurring being greater than or equal to two identical Feature Words, is determined as Similar content, merges automatically
Processing.
Variable cycle sampling in the step (2) refers to: continuous cycles sampling is carried out to the temperature of basic temperature event,
And speed situation is changed according to the temperature of basic temperature event in sampling and adjusts sampling period length.
The temperature of event in the step (1) is calculated by temperature calculation formula, temperature calculation formula: r=p* β+
q*γ+z*δ+s*ε；Wherein, p indicates that comment number, β indicate the comment factor, and q indicates to read number, and γ indicates to read the factor, and z is indicated
Forwarding number, δ indicate the forwarding factor, and s indicates to constitute event number, and ε indicates event factor；
After carrying out variable cycle sampling to the temperature of basic temperature event in the step (2), obtaining temperature value set is r
={ r1, r2, r3, r4 ... rn }, corresponding sampling time point set are t={ t1, t2, t3, t4 ... tn }, pass through temperature
Value set r and time point set t, first according to speed calculation formulaAnd acceleration formulaCalculate separately out the sets of speeds v={ v1, v2, v3, v4 ... vn } and acceleration set of temperature variation
A=a1, a2, a3, a4 ... an }；Again respectively by sets of speeds v and acceleration set a according to following formula, every 5 continuous
Period calculates differential weights mean value V={ V1, V2, V3, V4 ... Vn } and A={ A1, A2, A3, A4 ... An },
Differential weights mean value V formula:
Differential weights mean value A formula:
Sampling period T={ the T of the variable cycle sampling1,T2,T3,T4,...Tn, sampling period formula are as follows: T_
Default/ (1+ (V-T_Thd)/T_Thd), T_default is constant, indicates minimum period value；T_Thd is constant, is indicated most
Large period value；
Finally, calculating whether simultaneously fundamentals of forecasting temperature event has the possibility as mass incident according to following predictor formula
And it is likely to become the arrival time of mass incident,
Predictor formula:Wherein, P is constant, indicates the scale of basic temperature event
Desired value, t are the arrival time for being likely to become mass incident.
Using the present invention has the advantages that
1, event development is mapped as variable motion by the present invention, with the possible development scale of kinetic characteristics predicted events,
Calculability and arrival time can carry out more directly effective measuring and calculating in real time to event scale, and more efficient, precision is more
It is quasi-.Further, since the present invention can carry out more accurately prediction in real time to event scale in advance, such that in advance
The development of event is judged, in order to guide event benign development in optimal Public-opinion directing period, is effectively prevented
Event malignant development.
2, the present invention does not depend on supervised learning process, and no lag period can highly adapt to over, is now and following each
The new method of the class event rule of development.
3, the present invention is fully automated movement, is not necessarily to human intervention, reduces artificial and cost, uninterrupted daily from internet
Data are acquired, basic temperature event is extracted by information agglomerative clustering, then sampled by the intelligent alterable period, calculates early warning
May and the time.
Detailed description of the invention
Specific embodiment
The invention discloses a kind of methods for predicting event scale in social media by kinetic characteristics, and that uses is original
Data be social media propagate properties collection, lteral data include text, comment, numeric data include read number, forwarding number,
Number is commented on, specifically includes the following steps:
(1) unsupervised merging is carried out to social media content, generates event set, then calculate the heat of each event in event set
Degree finally extracts the basic temperature event for meeting basic temperature thresholding；Wherein, the temperature of event by reading number, comment number, turn
Hair number weighted comprehensive is calculated；
(2) variable cycle sampling is carried out to the temperature of basic temperature event, is calculated after sampling by variable motion rule
The temperature of basic temperature event is in accelerator or moderating process, if the temperature of basic temperature event is in accelerator,
Whether there is the possibility as mass incident by the prediction of result of the variable motion basis temperature event and is likely to become
The arrival time of mass incident.
Unsupervised merging in the step (1) refers to: first segmenting to social media content, and passes through TF-IDF
Feature Words extraction is carried out, when occurring being greater than or equal to two identical Feature Words, is determined as Similar content, merges automatically
Processing.
Variable cycle sampling in the step (2) refers to: continuous cycles sampling is carried out to the temperature of basic temperature event,
And speed situation is changed according to the temperature of basic temperature event in sampling and adjusts sampling period length.
The temperature of event in the step (1) is calculated by temperature calculation formula, temperature calculation formula: r=p* β+
q*γ+z*δ+s*ε；Wherein, p indicates that comment number, β indicate the comment factor, and q indicates to read number, and γ indicates to read the factor, and z is indicated
Forwarding number, δ indicate the forwarding factor, and s indicates to constitute event number, and ε indicates event factor；
After carrying out variable cycle sampling to the temperature of basic temperature event in the step (2), obtaining temperature value set is r
={ r1, r2, r3, r4 ... rn }, corresponding sampling time point set are t={ t1, t2, t3, t4 ... tn }, pass through temperature
Value set r and time point set t, by following process can fundamentals of forecasting temperature event whether have as mass incident can
It can and be likely to become the arrival time of mass incident.
Firstly, according to speed calculation formulaAnd acceleration formulaRespectively
Calculate temperature variation sets of speeds v={ v1, v2, v3, v4 ... vn } and acceleration set a=a1, a2, a3,
a4,...an}。
Secondly, respectively by sets of speeds v and acceleration set a according to following formula, every 5 continuous cycles are calculated
Power mean value V=V1, V2, V3, V4 ... Vn } and A=A1, A2, A3, A4 ... An }, differential weights mean value V formula:
Differential weights mean value A formula:
Wherein, since the development of event is continuous, and periodic sampling point can only represent the transient state at the moment, so needing
In conjunction with historical trend, preferably to reflect objective reality situation instantly.It can be used 5 periods when therefore implementing to adopt
Sample value carries out differential weights average value processing.
Sampling period T={ the T of the variable cycle sampling1,T2,T3,T4,...Tn, sampling period formula are as follows: T_
Default/ (1+ (V-T_Thd)/T_Thd), T_default is constant, indicates minimum period value；T_Thd is constant, is indicated most
Large period value.Wherein, it when event development and change are very fast, takes and shortens the sampling period so as to more dense sampling change procedure；
On the contrary, the sampling period can be extended when event development and change are slower.Variable cycle compares traditional fixed cycle sample mode, excellent
Point is more economical and efficient
Finally, calculating whether simultaneously fundamentals of forecasting temperature event has the possibility as mass incident according to following predictor formula
And it is likely to become the arrival time of mass incident,
Predictor formula:Wherein, P is constant, indicates the scale of basic temperature event
Desired value, t are the arrival time for being likely to become mass incident.
" formal employee " is become using one lovely cat of certain factory below and is had by network attention to above-mentioned steps
Body explanation, specific as follows:
(1) by participle, extraction keyword, following 4 information is found out, and event keyword is generated based on following information:
Wandering, trouble-shooter, conciliation, thigh, popularity, is good at, mouse at kitten；
A has one special " employee ", it not only the mouse in factory is suffered from tidied up it is clean, also become in spy
Different trouble-shooter.Such achievement is held, factory director gives special approval to that this mew has switched to " formal employee ", and the welfare for enjoying " ad eundem " waits for
It meets！The photo of mew has been sticked below employee's photo wall.
B has a vagrant cat recently, sprouts because being good to kill rats and sell, red at net in certain factory, reconciles someone in room every time
Ignition sound is big, and kitten just rushes in hugging thigh.
C, not long ago, a mew star people " little Huang " of certain factory are successfully loaded into employee manual.In the past, little Huang was a wandering
Mew, because the mouse of Chang Li archive office is too many, staff is just held back to it to catch mouse.But other is not only to grab mouse
Ability is strong, and more severe is regulating power.
D, [vagrant cat grabs mouse because arrogating to oneself, and sells to sprout by an armful thigh and mediate an issue] cat little Huang is the undisputed popularity of certain factory
King.Originally outer wandering it, because being good at mousing and sell and sprout, be in factory oneself seek a seat position.Factory director introduces, little Huang master
It is responsible for grabbing mouse to protect archives safety, weekend, it strolled with regard in kitchen itself, has grabbed mouse and has not also eaten up.Equal Mondays we
When opening the door, little Huang can be seen on doorway etc., it is put in side this two days mouse grabbed.In addition, little Huang, which is also good at, embraces thigh
Atmosphere is reconciled to alleviate.Once someone is under fire, sound is big, it just rushes in hugging thigh, and the people of regeneration gas is through so one
It embraces and also cools down.Since it takes up an official post, conciliation working efficiency is significantly improved in factory.There is kitchen in institute, everybody has a meal it just every time
It lies prone at table bottom, has some fish head, leftovers.Employee also can actively give small yellowish leukorrhea cat food in factory, other unit employee also can out of admiration for a famous person from the point of view of
Little Huang is eaten to its band.
It is again 654.0 by the temperature that temperature calculation formula (r=p* β+q* γ+z* δ+s* ε) calculates the event, it should
The temperature of event meets basic temperature thresholding, therefore extracts as basic temperature event.
(2) variable cycle sampling is carried out to the basis temperature event, basis is calculated by variable motion rule after sampling
If the temperature of the basic temperature event of the velocity and acceleration of temperature event is in accelerator, pre- by the result of variable motion
Survey whether the basis temperature event has the possibility as mass incident and be likely to become the arrival time of mass incident.
Specifically it see the table below:
As seen from the above table:
1, serial number 1-4 is the initial samples stage, is sampled at this time according to the highest minor cycle.
2, serial number 5-10 is online peak period, but progresses into the time of having a rest, and speed is gradually reduced, while predicting residue
Time t also extends therewith.
3, serial number 11-13 is the late into the night to early morning, and network liveness is very low at this time, and prediction remaining time t is unreachable.
4, serial number 14-19 is incident second day, event continuous diffusive transport in a network, and prediction remaining time t is gradually
Shorten, triggered extensive thresholding when being 0.
Event development is mapped as variable motion by the present invention, with the possible development scale of kinetic characteristics predicted events, meter
Possibility and arrival time are calculated, more directly effective measuring and calculating in real time can be carried out to event scale in advance, just and in advance to thing
The development of part judges, and is conducive to guide event benign development in optimal Public-opinion directing period, effectively prevents event
Malignant development.
Claims (4)
1. a kind of method for predicting event scale in social media by kinetic characteristics, it is characterised in that the following steps are included:
(1) unsupervised merging is carried out to social media content, generates event set, then calculate the temperature of each event in event set, most
The basic temperature event for meeting basic temperature thresholding is extracted afterwards；Wherein, the temperature of event is by reading number, comment number, forwarding number
Weighted comprehensive is calculated；
(2) variable cycle sampling is carried out to the temperature of basic temperature event, basis is calculated by variable motion rule after sampling
The temperature of temperature event is in accelerator or moderating process, if the temperature of basic temperature event is in accelerator, passes through
Whether the prediction of result of the variable motion basis temperature event has the possibility as mass incident and is likely to become big rule
The arrival time of mould event.
2. a kind of method for predicting event scale in social media by kinetic characteristics as described in claim 1, feature
Be: the unsupervised merging in the step (1) refers to: first segmenting to social media content, and is carried out by TF-IDF
Feature Words extract, and when occurring being greater than or equal to two identical Feature Words, are determined as Similar content, merge place automatically
Reason.
3. a kind of method for predicting event scale in social media by kinetic characteristics as described in claim 1, feature
Be: the variable cycle sampling in the step (2) refers to: continuous cycles sampling is carried out to the temperature of basic temperature event, and
Speed situation is changed according to the temperature of basic temperature event in sampling and adjusts sampling period length.
4. a kind of by event scale in kinetic characteristics prediction social media as described in any one of claim 1-3
Method, it is characterised in that: the temperature of the event in the step (1) is calculated by temperature calculation formula, temperature calculation formula:
R=p* β+q* γ+z* δ+s* ε；Wherein, p indicate comment number, β indicate comment the factor, q indicate read number, γ indicate read because
Son, z indicate forwarding number, and δ indicates the forwarding factor, and s indicates to constitute event number, and ε indicates event factor；
After carrying out variable cycle sampling to the temperature of basic temperature event in the step (2), obtaining temperature value set is r=
{ r1, r2, r3, r4 ... rn }, corresponding sampling time point set are t={ t1, t2, t3, t4 ... tn }, pass through temperature value set
R and time point set t, first according to speed calculation formulaAnd acceleration formula
Calculate separately out temperature variation sets of speeds v={ v1, v2, v3, v4 ... vn } and acceleration set a=a1, a2, a3,
a4,...an}；Again respectively by sets of speeds v and acceleration set a according to following formula, every 5 continuous cycles are calculated
Power mean value V=V1, V2, V3, V4 ... Vn } and A=A1, A2, A3, A4 ... An },
Differential weights mean value V formula:
Differential weights mean value A formula:
Sampling period T={ the T of the variable cycle sampling1,T2,T3,T4,...Tn, sampling period formula are as follows: T_
Default/ (1+ (V-T_Thd)/T_Thd), T_default is constant, indicates minimum period value；T_Thd is constant, is indicated most
Large period value；
Finally, according to following predictor formula calculate and fundamentals of forecasting temperature event whether have as mass incident possibility and
It is likely to become the arrival time of mass incident,
Predictor formula:Wherein, P is constant, indicates that the scale of basic temperature event is expected
Value, t is the arrival time for being likely to become mass incident.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811359804.5A CN109472415B (en) | 2018-11-15 | 2018-11-15 | Method for predicting event scale in social media through dynamic characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811359804.5A CN109472415B (en) | 2018-11-15 | 2018-11-15 | Method for predicting event scale in social media through dynamic characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109472415A true CN109472415A (en) | 2019-03-15 |
| CN109472415B CN109472415B (en) | 2021-11-19 |
Family
ID=65673710
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811359804.5A Active CN109472415B (en) | 2018-11-15 | 2018-11-15 | Method for predicting event scale in social media through dynamic characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109472415B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102937960A (en) * | 2012-09-06 | 2013-02-20 | 北京邮电大学 | Device and method for identifying and evaluating emergency hot topic |
| CN104035960A (en) * | 2014-05-08 | 2014-09-10 | 东莞市巨细信息科技有限公司 | Internet information hotspot predicting method |
| US9264391B2 (en) * | 2012-11-01 | 2016-02-16 | Salesforce.Com, Inc. | Computer implemented methods and apparatus for providing near real-time predicted engagement level feedback to a user composing a social media message |
| CN106294333A (en) * | 2015-05-11 | 2017-01-04 | 国家计算机网络与信息安全管理中心 | A kind of microblogging burst topic detection method and device |
| CN106503209A (en) * | 2016-10-26 | 2017-03-15 | Tcl集团股份有限公司 | A kind of topic temperature Forecasting Methodology and system |
| CN107798027A (en) * | 2016-09-06 | 2018-03-13 | 腾讯科技（深圳）有限公司 | A kind of heatrate Forecasting Methodology, information recommendation method and device |
| CN108549957A (en) * | 2018-04-11 | 2018-09-18 | 中译语通科技股份有限公司 | Internet topic trend auxiliary prediction technique and system, information data processing terminal |
-
2018
- 2018-11-15 CN CN201811359804.5A patent/CN109472415B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102937960A (en) * | 2012-09-06 | 2013-02-20 | 北京邮电大学 | Device and method for identifying and evaluating emergency hot topic |
| US9264391B2 (en) * | 2012-11-01 | 2016-02-16 | Salesforce.Com, Inc. | Computer implemented methods and apparatus for providing near real-time predicted engagement level feedback to a user composing a social media message |
| CN104035960A (en) * | 2014-05-08 | 2014-09-10 | 东莞市巨细信息科技有限公司 | Internet information hotspot predicting method |
| CN106294333A (en) * | 2015-05-11 | 2017-01-04 | 国家计算机网络与信息安全管理中心 | A kind of microblogging burst topic detection method and device |
| CN107798027A (en) * | 2016-09-06 | 2018-03-13 | 腾讯科技（深圳）有限公司 | A kind of heatrate Forecasting Methodology, information recommendation method and device |
| CN106503209A (en) * | 2016-10-26 | 2017-03-15 | Tcl集团股份有限公司 | A kind of topic temperature Forecasting Methodology and system |
| CN108549957A (en) * | 2018-04-11 | 2018-09-18 | 中译语通科技股份有限公司 | Internet topic trend auxiliary prediction technique and system, information data processing terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109472415B (en) | 2021-11-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Stukel et al. | Mesoscale ocean fronts enhance carbon export due to gravitational sinking and subduction | |
| Bianchi et al. | Global patterns of diel vertical migration times and velocities from acoustic data | |
| CN103236127B (en) | A kind of fiber fence system for monitoring intrusion and mode identification method thereof | |
| CN105718598A (en) | AT based time model construction method and network emergency early warning method | |
| CN103886344A (en) | Image type fire flame identification method | |
| CN102222283A (en) | Data management system for face recognition | |
| Cerco et al. | Management modeling of suspended solids in the Chesapeake Bay, USA | |
| Sevadjian et al. | Shoreward advection of phytoplankton and vertical re-distribution of zooplankton by episodic near-bottom water pulses on an insular shelf: Oahu, Hawaii | |
| CN109918649A (en) | A kind of suicide Risk Identification Method based on microblogging text | |
| Lu et al. | Inner-core sea surface cooling induced by a tropical cyclone | |
| CN109472415A (en) | A method of event scale in social media is predicted by kinetic characteristics | |
| Zhang et al. | Spatial variability in plankton biomass and hydrographic variables along an axial transect in Chesapeake Bay | |
| Gerges et al. | A novel Bayesian deep learning approach to the downscaling of wind speed with uncertainty quantification | |
| CN202230509U (en) | Face identification data management system | |
| Edburg et al. | A simple model to predict scalar dispersion within a successively thinned loblolly pine canopy | |
| Beck et al. | Estimating occupancy to monitor Northern Goshawk in the central Rocky Mountains | |
| Perry et al. | Environmental influences on the availability of smooth pink shrimp, Pandalus jordani, to commercial fishing gear off Vancouver Island, Canada | |
| Li et al. | Intelligent fuzzy optimization algorithm of data mining based on BP neural network | |
| Al-Roby et al. | Data mining techniques for wind speed analysis | |
| CN114422193A (en) | Botnet risk assessment method and device | |
| Mcleay et al. | A biophysical model to assess the trade‐off between larval recruitment and catch in southern A ustralia's largest prawn fishery | |
| Casula et al. | Local abundance and observer's identity affect visual detectability of Sardinian mountain newts. | |
| Bolla et al. | Weather Forecasting Method from Sensor Transmitted Data for Smart Cities Using IoT | |
| Li et al. | Intelligent fire monitoring system based on the information fusion algorithm | |
| Wu et al. | A convolutional-LSTM model for nitrogen oxide emission forecasting in FCC unit |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |