CN101819585A - Device and method for constructing forum event dissemination pattern - Google Patents
Device and method for constructing forum event dissemination pattern Download PDFInfo
- Publication number
- CN101819585A CN101819585A CN201010133617A CN201010133617A CN101819585A CN 101819585 A CN101819585 A CN 101819585A CN 201010133617 A CN201010133617 A CN 201010133617A CN 201010133617 A CN201010133617 A CN 201010133617A CN 101819585 A CN101819585 A CN 101819585A
- Authority
- CN
- China
- Prior art keywords
- forum
- event
- propagation
- network
- incident
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a device and a method for constructing a forum event dissemination pattern. The device for constructing the forum event dissemination pattern is formed by connecting a network event data collecting device, a network event data pre-processing device, a hot event finding device, a dissemination pattern constructing device and an output device. The method for constructing the forum event dissemination pattern comprises the following steps of: acquiring network event information; finding a network event; initializing the dissemination pattern; generating the dissemination pattern; calculating weight of the forum dissemination; and outputting the event dissemination pattern. By the method and the device, hot event dissemination process and route information published at the forum can be found in real time, and contribution of each forum to the event dissemination can be known, so that a network manager can know and monitor the trend of the network hot event more completely. The device and the method can be widely applied to the application field of network public sentiment management, internet intelligent information processing and the like.
Description
Technical field
The present invention relates to utilize the technology of information intelligent analysis of computer technology auxiliary network or public sentiment management, it is particularly related to and utilizes natural language processing and data mining technology to find the propagation of focus incident in the forum, the apparatus and method of diffusion process.
Background technology
BBS has become the platform that the netizen delivers viewpoint and raps off.Because the anonymity that has of BBS, opening, characteristics such as interactive make that the netizen is free to express an opinion, simultaneously, the widespread use of BBS has brought new challenge for the public sentiment supervision and oversight.The information that often makes rapidly that BBS information is propagated becomes uncontrollable, a course of event netizen's reprinting, propagation, just can on network, spread rapidly, finally may break out public opinion, therefore need find in real time and follow the tracks of the communication process of focus incident, in time the up-to-date trend of grasp incident development, thereby carry out the network public-opinion monitoring better.Find between the forum that the focus incident communication process is by automatic extracting, the topic detection of target forum information are found the communication process of focus incident between different forums, reach the purpose that its source was accurately monitored and analyzed in to network focus event implementation.The basic ideas that the focus incident communication process is found between the forum are to adopt text mining technical finesse forum information data.
At present obtaining certain achievement in research aspect the discovery of network focus event propagation process both at home and abroad.Some scholar starts with from the research social relation network, analyzes the communication process of topic in forum.Utilize statistical regression model to investigate of the influence of each factor to propagating
[1], this method has illustrated that mainly propagation has important contribution to the leader of opinion of forum to topic.Along with the development of complex network research and deeply, some scholar finds that the propagation of the rumor of real world also has the characteristic of worldlet network and no scale network
[2], the event propagation of this and BBS has something in common.People such as Wan Xiaojun have studied the propagation of information between each news website and have reprinted process
[3]At first an all document of media event is all obtained,, calculate the source document of each document successively according to propagating keyword and similarity, thereby obtain the whole communication process of this topic between different web sites by the issuing time ordering.Zhao Li has proposed to have paroxysmal topic propagation model in the blog network
[4]Liveness and outside field intensity according to each blog node are calculated the probability that certain blog node is paid close attention to topic, thereby prediction may participate in the blog node of topic, though the research object of the document is the blog network, achievement in research is propagated forum's topic the certain experiences meaning.
Analyze existing forum focus incident communication process discovery technique as can be seen, realize that the technology path that the focus incident communication process is found mainly contains two:
1. analyze the communication process of focus incident in the subordinate act: comprising the interactive relation between the user between the model in the forum and its each answer, therefore can come the propagation of topic is studied from analyzing interpersonal relational matrix.
2. analyze the communication process of focus incident from content, mainly its travel path in forum is analyzed in the reprinting of an incident, describe the whole travel path and the process of incident intuitively according to each big website.
Propagation to topic in the forum does not at present also have how deep research, and the focus incident of relative news, blog is propagated certain research basis.
The open file relevant with the present patent application has:
[1] palace brightness, Xu Yu. the influence factor [J] that efficient BBS social structure and information are propagated. XI AN JIAOTONG UNIVERSITY Subject Index (social science version) .2007 (01);
[2] all brightness. the worldlet network characteristic research [J] that rumor is propagated. the journal .2005 of Wuhan University Of Science and Engineering (01);
[3]Wan，X.and?J.Yang.Learning?information?diffusion?process?on?theweb.in?Proceedings?of?the?16th?international?conference?on?World?WideWeb.2007.Banff，Alberta，Canada：ACM；
[4] Zhao Li, Yuan Rui is amiable and compliant, Guan Xiaohong. have paroxysmal topic propagation model [J] in the blog network. and software journal .2009 (20).
Summary of the invention
The object of the present invention is to provide a kind of construction device and construction method that the forum event dissemination pattern of technical support can be provided for network intelligence information processing and public sentiment analysis.
The object of the present invention is achieved like this:
The construction device of forum event dissemination pattern of the present invention finds that by network event transacter, network event data pretreatment unit, focus incident device, propagation figure construction device and output unit connect and compose; The network event transacter is collected the primary data information (pdi) fan-in network event data pretreatment unit relevant with certain incident; Network event data pretreatment unit is handled the primitive event data that the network event transacter is collected, the primitive event data are converted to the time of delivering, place forum, the model content of text that representation that the construction device that is fit to this forum event dissemination pattern handles promptly extracts every piece of model, and the model content of text carried out participle, feature extraction, the model content of text shows with the form of VSM the most at last; Focus incident finds that device carries out cluster to the pretreated data of network event data, and the related data of same incident is brought together; Propagation figure construction device is set up the route of transmission of incident between each forum, and calculates the contribution margin of each forum to event propagation, obtains propagating initial forum, propagates core forum; The details of the communication process of the focus incident that output unit output is found be included in the route of transmission between each forum, and each forum are to the contribution margin of event propagation.
The construction device of forum event dissemination pattern of the present invention can also comprise:
1, described network event data pretreatment unit is connected to form by network event data characteristics extraction unit, feature weight computing unit; Network event data characteristics extraction unit can be expressed the content of network event data and the feature extraction of suitable system handles is come out; The feature weight computing unit, the weighted value of employing TF-IDF method calculated characteristics is for similarity between the calculating incident provides support.
2, described propagation figure construction device is formed by propagating figure initialization unit, propagation figure generation unit and forum's propagation weight calculation unit; Propagation figure initialization unit is carried out according to time sequence the data that belong to an incident, determine to propagate initial forum, and prepares for subsequent analysis calculating; Propagation figure generation unit is set up two propagation contacts between the forum according to this forum priority principle and similarity principle, thereby obtains the route of transmission of incident in all forums; Forum propagates weight calculation unit and calculates the contribution of each forum to event propagation, thereby obtains the core forum of communication events.
The construction method of forum event dissemination pattern of the present invention may further comprise the steps:
Network event information acquisition step: obtain the current forum's model data that are published on each World Jam, describe event content;
Network event is found step: adopt the Single-pass incremental clustering algorithm that the forum's model data that collect are handled, find the some network events that exist in the current forum;
Propagate the figure initialization step: to notice's data of belonging to same incident by delivering the time order and function rank order, and extract every piece of model the ground domain information, determine to propagate initial forum, calculate for subsequent analysis and prepare;
Propagate figure and generate step: set up two propagation contacts between the forum according to this forum priority principle and similarity principle, thereby obtain the route of transmission of incident in all forums;
Forum's dissemination right re-computation step: calculate the contribution of each forum to event propagation, the weighted sum of each forum's node out-degree promptly is the propagation contribution margin of this forum among the propagation figure, presses the ordering of contribution margin size, thereby obtains the core forum of communication events;
Event dissemination pattern output step: be used to export the network focus event propagation approach information of expressing with the digraph form, comprise the initial forum of communication events, the weights of core forum, forum and route of transmission information.
The process that propagation figure of the present invention sets up is based upon on following two principle basis:
Principle 1: this forum priority principle: when post when a plurality of in the candidate source of one piece of model, the preferential selection belongs to the source card of the candidate source card of same forum as this model with this model;
Principle 2: similarity principle: post a plurality ofly when the candidate source of one piece of model, and when belonging to same forum with this model, select the source of this model of conduct of similarity maximum to paste without any candidate source card.
Beneficial effect of the present invention is, can find to issue at present focus incident communication process and approach information in the forum in real time by the present invention, and can understand the contribution that each forum does event propagation, can understand situation with the monitor network focus incident more all sidedly to help the network management personnel.
The present invention can be widely used in applications such as network public-opinion management, internet intelligent information processing.
Description of drawings
Accompanying drawing 1 apparatus structure block diagram of the present invention;
The detail flowchart of 2 one network focuses of accompanying drawing event propagation process discover method;
The typical applied environment that accompanying drawing 3 carries out an invention;
Accompanying drawing 4 network focus event propagation processes represent.
Embodiment
For example the present invention is done description in more detail below in conjunction with accompanying drawing:
Figure 1 shows that a kind of enforcement multi-angle network event focus finds system, comprising:
The network event transacter is used to import the original contents information relevant with certain incident;
Network event data pretreatment unit: be used to handle the primitive event data that the network event transacter is collected, be converted into the representation that is fit to the native system processing;
Focus incident is found device: be used for the pretreated data of network event data are carried out cluster, the related data of same incident is brought together;
Propagate the figure construction device: be used to the route of transmission of the incident of setting up between each forum, and calculate the contribution margin of each forum, obtain propagating initial forum, propagate core forum to event propagation;
Output unit: be used for the details of the communication process of the focus incident that output system finds, be included in the route of transmission between each forum, and each forum be to the contribution margin of event propagation.
Described network event transacter can be a web crawlers, also can be to be realized by manual by the keeper.
Described output unit can output to the route of transmission information of the network focus incident found in other the signal conditioning package and memory storage such as personal computer.
Fig. 2 has provided the principle flow chart of forum's focus incident communication process discover method.
1. network event data aggregation
The present invention utilizes web crawlers to gather each World Jam model data with incremental mode.Concrete gatherer process is based on breadth-first strategy, and the acquiescence sampling depth is three layers.Under this strategy, reptile is at first gathered the forum's model raw data that realizes specifying in kind of the subpage frame, gathers the related data in the peer link page pointed in this page then, by that analogy, finishes until whole collections.
2. network event data pre-service
The present invention adopts the formalized description of vector space model as network event, and the network event data vectorization comprises the steps:
(1) utilizes dictionary for word segmentation that the network event data are carried out word segmentation processing, extract notional word wherein, remove function word and stop words;
(2) adopt the TF-IDF method to determine the weight of each speech behind the participle, the computing method of TF-IDF as shown in the formula:
Wherein: W
iThe weight of representing i feature speech, TF
i(N represents total number of documents for t, the d) frequency of occurrences of expression speech t in document d, and DF (t) expression comprises the number of files of speech t.
(3) by the weight of each speech as component, form the vector representation of this network event.
3. focus incident is found
Each class of the present invention is described with a vector space model, after the main subsides information of follow-up model carried out vectorization, calculate the similarity of itself and each classification, and the size according to similarity is sorted out: if similarity is less than new events threshold value (being assumed to be 0.3), be the newly-built class of this model, otherwise, this model is classified as the classification of similarity maximum.Similarity is calculated the formula below adopting:
Wherein D and T are two vectors, q
iBe the value of i the component of vectorial D, d
iBe the value of i component of vector T, H is the dimension of vector.
4. propagating figure makes up
The published method of model is divided into three types in the forum: the firstth, and former subsides, i.e. first of each incident piece of model.For each topic, has only one piece of former subsides; The secondth, be subjected to other model influences of this forum and the model of issue, the 3rd is to be subjected to other forum's influences and the model of issue.D=＜d
1, d
2, d
3..., d
n＞for to deliver the time by the ordering after arriving first by model, d
1Be the former subsides of this topic; For each d
i(1＜i＜=n), analyze itself and the relation that comes the model of its front: if having j, (1=＜j＜i), make d
jWith d
iBelong to same forum, then think d
iIssued by the influence of other models of this forum, then set up a limit of pointing to self in this forum, the weights on limit add 1; If any one d
j(1=＜j＜i) all with d
iDo not belong to same forum, then calculate d
iWith d
j(text similarity of 1=＜j＜i) is selected the dj of similarity maximum, and the di influence that is subjected to dj place forum is issued so, sets up one by d
jPlace forum points to d
iThe limit of place forum, weights are 1.According to above-mentioned influence relation, can set up a complete propagation figure of forum.On the forum propagation figure basis of setting up, can calculate the contribution margin of each forum to propagating.The influence power of a forum is the weighted sum of its out-degree.By the influence power size is sorted, can draw the big forum of contribution, i.e. core forum of Chuan Boing propagated in topic.The method that adopts when wherein the similarity calculating method of Cai Yonging is found with focus incident is identical.
5. embodiment scene and result describe
In order to verify validity of the present invention, we have built typical applied environment, and Fig. 3 is the typical applied environment of embodiment.The dawn server of AMD OPTERON 2G is adopted in experiment, and operating system is the Linux enterprise version of 2.6.16.19 kernel.
Experiment has been collected model about incident " student of ZhengFa University cuts professor to death " from 7 forums, and 7 pieces of models are collected by each forum.
Fig. 4 has represented the communication process of incident.
As can be known from Fig. 4, the initial forum that propagates this topic is a forum of Xinhua, and as can be known each forum to the contribution of this topic: forum of Xinhua: 11, Chinese forum: 7, all the other forums: 6.Therefore, forum of Xinhua and Chinese forum are bigger to the propagation contribution of this topic, are the core forums that propagates topic.From figure,, can also recognize the relation of influencing each other between the forum intuitively except understanding the core forum that propagates certain topic.
Claims (4)
1. the construction device of a forum event dissemination pattern is characterized in that: find that by network event transacter, network event data pretreatment unit, focus incident device, propagation figure construction device and output unit connect and compose; The network event transacter is collected the primary data information (pdi) fan-in network event data pretreatment unit relevant with certain incident; Network event data pretreatment unit is handled the primitive event data that the network event transacter is collected, the primitive event data are converted to the representation of the construction device processing that is fit to this forum event dissemination pattern, promptly extract the time of delivering, place forum, the model content of text of every piece of model, and the model content of text carried out participle, feature extraction, the model content of text shows with the form of vector space model the most at last; Focus incident finds that device carries out cluster to the pretreated data of network event data, and the related data of same incident is brought together; Propagation figure construction device is set up the route of transmission of incident between each forum, and calculates the contribution margin of each forum to event propagation, obtains propagating initial forum, propagates core forum; The details of the communication process of the focus incident that output unit output is found be included in the route of transmission between each forum, and each forum are to the contribution margin of event propagation.
2. the construction device of forum event dissemination pattern according to claim 1, it is characterized in that: described network event data pretreatment unit is connected to form by network event data characteristics extraction unit, feature weight computing unit; Network event data characteristics extraction unit can be expressed the content of network event data and the feature extraction of suitable system handles is come out; The feature weight computing unit, the weighted value of employing TF-IDF method calculated characteristics is for similarity between the calculating incident provides support.
3. the construction device of forum event dissemination pattern according to claim 1 and 2 is characterized in that: described propagation figure construction device is by propagating the figure initialization unit, propagating figure generation unit and forum and propagate weight calculation unit and form; Propagation figure initialization unit is carried out according to time sequence the data that belong to an incident, determine to propagate initial forum, and prepares for subsequent analysis calculating; Propagation figure generation unit is set up two propagation contacts between the forum according to this forum priority principle and similarity principle, thereby obtains the route of transmission of incident in all forums; Forum propagates weight calculation unit and calculates the contribution of each forum to event propagation, thereby obtains the core forum of communication events.
4. the construction method of a forum event dissemination pattern is characterized in that may further comprise the steps:
Network event information acquisition step: obtain the current forum's model data that are published on each World Jam, describe event content;
Network event is found step: adopt the Single-pass incremental clustering algorithm that the forum's model data that collect are handled, find the some network events that exist in the current forum;
Propagate the figure initialization step: to notice's data of belonging to same incident by delivering the time order and function rank order, and extract every piece of model the ground domain information, determine to propagate initial forum, calculate for subsequent analysis and prepare;
Propagate figure and generate step: set up two propagation contacts between the forum according to this forum priority principle and similarity principle, thereby obtain the route of transmission of incident in all forums;
Forum's dissemination right re-computation step: calculate the contribution of each forum to event propagation, the weighted sum of each forum's node out-degree promptly is the propagation contribution margin of this forum among the propagation figure, presses the ordering of contribution margin size, thereby obtains the core forum of communication events;
Event dissemination pattern output step: be used to export the network focus event propagation approach information of expressing with the digraph form, comprise the initial forum of communication events, the weights of core forum, forum and route of transmission information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010133617A CN101819585A (en) | 2010-03-29 | 2010-03-29 | Device and method for constructing forum event dissemination pattern |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010133617A CN101819585A (en) | 2010-03-29 | 2010-03-29 | Device and method for constructing forum event dissemination pattern |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101819585A true CN101819585A (en) | 2010-09-01 |
Family
ID=42654687
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010133617A Pending CN101819585A (en) | 2010-03-29 | 2010-03-29 | Device and method for constructing forum event dissemination pattern |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101819585A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567407A (en) * | 2010-12-22 | 2012-07-11 | 北大方正集团有限公司 | Method and system for collecting forum reply increment |
| CN104537096A (en) * | 2015-01-09 | 2015-04-22 | 哈尔滨工程大学 | Microblog message influence measuring method based on microblog message propagation tree |
| CN105718590A (en) * | 2016-01-27 | 2016-06-29 | 福州大学 | Multi-tenant oriented SaaS public opinion monitoring system and method |
| CN106341265A (en) * | 2016-09-09 | 2017-01-18 | 百度在线网络技术（北京）有限公司 | Display method of network data and apparatus thereof |
| CN106383919A (en) * | 2016-11-21 | 2017-02-08 | 青岛农业大学 | Method and system for determining news transmission effect |
| CN106776841A (en) * | 2016-11-28 | 2017-05-31 | 福建亿榕信息技术有限公司 | The acquisition methods and system of a kind of internet public feelings event propagation index |
| CN106844786A (en) * | 2016-12-08 | 2017-06-13 | 中国电子科技网络信息安全有限公司 | A kind of public sentiment region focus based on text similarity finds method |
| CN106933840A (en) * | 2015-12-29 | 2017-07-07 | 北京国双科技有限公司 | Forum's catalogue page content crawling method and device |
| CN107436877A (en) * | 2016-05-25 | 2017-12-05 | 北京京东尚科信息技术有限公司 | Much-talked-about topic method for pushing and device |
| WO2018027392A1 (en) * | 2016-08-06 | 2018-02-15 | 柳柏宏 | Method for indicating technology ownership upon forwarding of forum information, and forwarding system |
| CN110472105A (en) * | 2019-08-06 | 2019-11-19 | 电子科技大学 | A kind of social networks event evolution method for tracing divided based on the time |
| CN111797333A (en) * | 2020-06-04 | 2020-10-20 | 南京擎盾信息科技有限公司 | Public opinion spreading task display method and device |
| CN116756233A (en) * | 2023-08-23 | 2023-09-15 | 博智安全科技股份有限公司 | Situation data processing method and device, electronic equipment and storage medium |
-
2010
- 2010-03-29 CN CN201010133617A patent/CN101819585A/en active Pending
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567407B (en) * | 2010-12-22 | 2014-07-16 | 北大方正集团有限公司 | Method and system for collecting forum reply increment |
| CN102567407A (en) * | 2010-12-22 | 2012-07-11 | 北大方正集团有限公司 | Method and system for collecting forum reply increment |
| CN104537096A (en) * | 2015-01-09 | 2015-04-22 | 哈尔滨工程大学 | Microblog message influence measuring method based on microblog message propagation tree |
| CN106933840A (en) * | 2015-12-29 | 2017-07-07 | 北京国双科技有限公司 | Forum's catalogue page content crawling method and device |
| CN105718590A (en) * | 2016-01-27 | 2016-06-29 | 福州大学 | Multi-tenant oriented SaaS public opinion monitoring system and method |
| CN107436877A (en) * | 2016-05-25 | 2017-12-05 | 北京京东尚科信息技术有限公司 | Much-talked-about topic method for pushing and device |
| WO2018027392A1 (en) * | 2016-08-06 | 2018-02-15 | 柳柏宏 | Method for indicating technology ownership upon forwarding of forum information, and forwarding system |
| CN106341265A (en) * | 2016-09-09 | 2017-01-18 | 百度在线网络技术（北京）有限公司 | Display method of network data and apparatus thereof |
| CN106341265B (en) * | 2016-09-09 | 2019-11-08 | 百度在线网络技术（北京）有限公司 | The methods of exhibiting and device of network data |
| CN106383919A (en) * | 2016-11-21 | 2017-02-08 | 青岛农业大学 | Method and system for determining news transmission effect |
| CN106776841A (en) * | 2016-11-28 | 2017-05-31 | 福建亿榕信息技术有限公司 | The acquisition methods and system of a kind of internet public feelings event propagation index |
| CN106844786A (en) * | 2016-12-08 | 2017-06-13 | 中国电子科技网络信息安全有限公司 | A kind of public sentiment region focus based on text similarity finds method |
| CN110472105A (en) * | 2019-08-06 | 2019-11-19 | 电子科技大学 | A kind of social networks event evolution method for tracing divided based on the time |
| CN111797333A (en) * | 2020-06-04 | 2020-10-20 | 南京擎盾信息科技有限公司 | Public opinion spreading task display method and device |
| CN111797333B (en) * | 2020-06-04 | 2021-04-20 | 南京擎盾信息科技有限公司 | Public opinion spreading task display method and device |
| CN116756233A (en) * | 2023-08-23 | 2023-09-15 | 博智安全科技股份有限公司 | Situation data processing method and device, electronic equipment and storage medium |
| CN116756233B (en) * | 2023-08-23 | 2023-11-07 | 博智安全科技股份有限公司 | Situation data processing method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101819585A (en) | Device and method for constructing forum event dissemination pattern | |
| Dahal et al. | Topic modeling and sentiment analysis of global climate change tweets | |
| CN112328801B (en) | Method for predicting group events by event knowledge graph | |
| Varshney et al. | A review on rumour prediction and veracity assessment in online social network | |
| Ruan et al. | Public perception of electric vehicles on reddit over the past decade | |
| CN103544255A (en) | Text semantic relativity based network public opinion information analysis method | |
| CN104536956A (en) | A Microblog platform based event visualization method and system | |
| CN105718587A (en) | Network content resource evaluation method and evaluation system | |
| Tang et al. | Social media-based disaster research: Development, trends, and obstacles | |
| Hsu et al. | Integrating machine learning and open data into social Chatbot for filtering information rumor | |
| CN103902621A (en) | Method and device for identifying network rumor | |
| CN114330966A (en) | Risk prediction method, device, equipment and readable storage medium | |
| Sun et al. | Socialwave: visual analysis of spatio-temporal diffusion of information on social media | |
| CN112395513A (en) | Public opinion transmission power analysis method | |
| Khun et al. | Visualization of Twitter sentiment during the period of US banned huawei | |
| Bai et al. | A rumor detection model incorporating propagation path contextual semantics and user information | |
| Al-Kfairy et al. | A sentiment analysis approach for identifying users’ security and privacy perception of metaverse in twitter | |
| CN111582757B (en) | Method, device, equipment and computer readable storage medium for analyzing fraud risk | |
| Singh et al. | Fighting for Information Credibility: AnEnd-to-End Framework to Identify FakeNews during Natural Disasters. | |
| CN109063485A (en) | A kind of vulnerability classification statistical system and method based on loophole platform | |
| Appling et al. | Deriving disaster-related information from social media | |
| CN105653649A (en) | Identification method and device of low-proportion information in mass texts | |
| Syliongka et al. | Combining automatic and manual approaches: Towards a framework for discovering themes in disaster-related Tweets | |
| Jones et al. | Topic modeling of small sequential documents: Proposed experiments for detecting terror attacks | |
| Jingjing et al. | Mining social influence in microblogging via tensor factorization approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Open date: 20100901 |