SlideShare a Scribd company logo
1 of 32
ISCRAM 2013, May 12-15 1
PhaseVis:
Visualizing the Four Phases of
Emergency Management
Through the Lens of Social
Media
Seungwon Yang et al.
Department of Computer Science, Virginia Tech
5/13/2013
Outline
1. Motivation
2. Hurricane Isaac
3. Approach (Selection, Classification, Visualization)
4. PhaseVis in Action
5. Limitations
6. Discussion
ISCRAM 2013, May 12-15 2
1. Motivation
 Four Phases of Emergency Management Model
 FEMA training material adds ‘Prevention/Protection’
http://training.fema.gov/EMIWeb/IS/IS230B/IS230bCourse.pdf
ISCRAM 2013, May 12-15 3
Response
Recovery
Mitigation
Preparedne
ss
2. Hurricane Isaac: Trajectory
ISCRAM 2013, May 12-15 4
8/24
Cuba, Hispaniola:
approx. 30 died
8/28-29
Mississippi River, Georgia,
Port Fourchon, LA: 9 died
After 8/30
N. Louisiana: degenerated
to tropical depression
*Image by Cyclonebiskit (Wikipedia)
8/21
Tropical storm
Isaac
8/19-20
Extratropical cyclone
ISCRAM 2013, May 12-15 5
Disaster Tweets
with emergency
orgs, agency names
Visualiza on
&
Interac on
Manual
Labeling
Training
Data
Trained
classifica on
model
Cleaned
Tweets
Original
Tweets
Original &
Retweets
Classified
Tweets
Select and
Preprocess
Tweets
Classify into
4 phases
Implement
visualization
& interaction
3. Overall Approach
 Tweet collection using ‘#isaac’ with
yourTwapperKeeper
 Situation report & Information sharing
 Majority of tweets
 Embedded URLs: news webpages, videos,
photographs
 Personal activity report
 Very few
ISCRAM 2013, May 12-15 6
3. Tweet Collection
 Approx. 56,000 English tweets collected with
‘#Isaac’
 5,677 tweets (10%) with reference to Red Cross,
FEMA, or Salvation Army
 1,453 non-retweets
 1,121 manually labeled with one of four phases
(response, recovery, mitigation, preparedness)
ISCRAM 2013, May 12-15 7
3. Building a Dataset (1/2)
 Tweet text + resource title
ISCRAM 2013, May 12-15 8
Nice article abt our Dir. Of emerg srvcs @leopratte
in #Louisiana organizing #redcross #Isaac relief
http://t.co/D4RPr33n
3. Building a Dataset (2/2)
ISCRAM 2013, May 12-15 9
Response More than 4,700 people in as many as 80 shelters in 7
states overnight; more than 3,000 #RedCross workers (37
from KC region) at #Isaac
Recovery FEMA announces that federal aid has been made available
for the state of Louisiana. #Isaac
Mitigation FEMA mitigations advisers to offer rebuilding tips in St.
Bernard and Ascension Parishes. http://t.co/ZziRGOGw
#Isaac
Preparednes
s
Very cool app! MT @redcross: Our hurricane app has info
on #RedCross shelters, a toolkit w flashlight, alarm
http://t.co/E7o1rtJK #Isaac
3. Examples of 4 Phases
 SVM multiclass with linear kernel
 Large num. of features, small num. of training
examples
 Naïve Bayes multinomial
 Bag-of-words model fits well for tweet data
 Random forest
 One of the robust algorithms for text classification
ISCRAM 2013, May 12-15 10
3. Classification Algorithms
 TF, normalization, stemming applied
 Tuned classifier, 10 fold cross-validation
ISCRAM 2013, May 12-15 11
Precision Weighted F
Measure
Naïve Bayes
multinomial
77.87% 0.782
Random forest 76.27% 0.754
SVM
multiclass
(linear kernel)
80.82% Reported slightly
lower than Naïve
Bayes multinomial
3. Classification Cross-Validation
ISCRAM 2013, May 12-15 12
3. Tweet Visualization
WHAT
WHEN
WHERE
WHO
 WHAT (Phases, List)
 Phases: ThemeRiver, D3 visualization toolkit
 Tweet List: JqGrid Library
 WHEN (Timeline)
 JavaScript
 WHERE (user locations)
 Google Maps API
 WHO (user mention network)
 Gephi graph format, Sigma.js
ISCRAM 2013, May 12-15 13
3. Visualization Implementation
ISCRAM 2013, May 12-15 14
4. PhaseVis in Action (8/23-8/24)
 Majority of tweets in Preparedness phase (84%)
 Content: fill up the gas tank, hurricane App,
preparedness tips, replace food/water in emergency
kit, etc…
 Clustered around
 Red Cross, FEMA, & CraigatFEMA
 Study focus was rather on the US (English tweets)
 Spanish tweets from Cuba, Hispaniola not
considered
 Unable to understand phases in such areas
ISCRAM 2013, May 12-15 15
4. Summary (8/23-8/24)
ISCRAM 2013, May 12-15 16
4. PhaseVis in Action (8/28-8/29)
- Mainly in Louisiana, Mississippi, Georgia -
 High increase in tweet volume
 Isaac landed in the US in 8/28 with hurricane
strength
 Response (20%), Recovery (34%), Mitigation (0%),
Preparedness (46%)
 Content:
 Recruiting volunteers (Response, Recovery)
 Asking for donations/support (Recovery)
 RT regarding ‘Mitt Romney’
 Providing shelters (Response)…
ISCRAM 2013, May 12-15 17
4. Tweet Details (8/28-8/29)
ISCRAM 2013, May 12-15 18
4. PhaseVis in Action (9/5-9/7)
- US continued -
 Mostly Recovery phase (75%), followed by
continued Response actions…
 Lots of activities in New Orleans, Baton Rouge,
Louisiana
 Active tweet account: FEMA, Red Cross,
RedCrossSELA (South East Louisiana)
ISCRAM 2013, May 12-15 19
4. Tweet Details (9/5-9/7)
ISCRAM 2013, May 12-15 20
5. Limitations
 Language
 Only English tweets considered
 Unable to analyze Spanish tweets when Isaac hit Cuba &
Hispaniola
 Small data set
 Only tweets containing FEMA, Red Cross & Salvation
Army
 E.g., RedCrossSELA, SalvationArmy, craigatFEMA, …
 Approx. 10% of tweets had those names
ISCRAM 2013, May 12-15 21
6. Discussion
 What are other valuable information to uncover from
disaster tweets and why are they important?
 Sentiment, Reliability of tweets
 Embedded URLs: news articles, images, videos…
 ??
 To what extent can tweet analysis actually help
emergency managers in the field?
 Identification of ‘actionable’ tweets from affected areas,
victims, and witnesses…
 ??
 NSF for funding: IIS-0916733 (CTRnet project)
 Internet Archive for collaboration
 Big thanks to co-authors who couldn’t come here
 Haeyong Chung, Xiao Lin, Sunshin Lee, Liangzhe
Chen, Andy Wood, and the CTRnet Team
ISCRAM 2013, May 12-15 22
Acknowledgment
Thank you!
Questions?
ISCRAM 2013, May 12-15 23
Supplementary
ISCRAM 2013, May 12-15 24
Evaluation
 Preprocessing & Accuracy
ISCRAM 2013, May 12-15 25
TF IDF Normali
zation
Naïve Bayes
Multinomial
SVM Multiclass
76% 80.1%
X 77% 80.4%
X 60% 78.8%
X X 78.1%
X 75% 80.4%
X X 78% 80.8%
X X 63% 78.9%
X X X 79.0%
ISCRAM 2013, May 12-15 26
3. Visualization: Phase View
ISCRAM 2013, May 12-15 27
Overview Detail
3. Visualization: Social Network View
ISCRAM 2013, May 12-15 28
3. Visualization: Location View
ISCRAM 2013, May 12-15 29
Is_R
(Retweet
check)
Tweet
Text
Phases Date
3. Visualization: Tweet View
Use Case & Demo
http://spare05.dlib.vt.edu/~ctrvis/phasevis/ind
ex_may.html
ISCRAM 2013, May 12-15 30
ISCRAM 2013, May 12-15 31
ISCRAM 2013, May 12-15 32

More Related Content

Similar to PhaseVis: What, When, Where, and Who in Visualizing the Four Phases of Emergency Management Through the Lens of Social Media

Rfs & social media
Rfs & social mediaRfs & social media
Rfs & social mediaBGTT_SYD
 
Crisis Information Processing - with the power of A.I.
Crisis Information Processing - with the power of A.I.Crisis Information Processing - with the power of A.I.
Crisis Information Processing - with the power of A.I.The Open University
 
05 june2013ezell uasi_nationalhls_2013
05 june2013ezell uasi_nationalhls_201305 june2013ezell uasi_nationalhls_2013
05 june2013ezell uasi_nationalhls_2013Barry Ezell
 
Helping Crisis Responders Find the Informative Needle in the Tweet Haystack
Helping Crisis Responders Find the Informative Needle in the Tweet HaystackHelping Crisis Responders Find the Informative Needle in the Tweet Haystack
Helping Crisis Responders Find the Informative Needle in the Tweet HaystackCOMRADES project
 
Utilizing Community Volunteered Information to Enhance Disaster Situational A...
Utilizing Community Volunteered Information to Enhance Disaster Situational A...Utilizing Community Volunteered Information to Enhance Disaster Situational A...
Utilizing Community Volunteered Information to Enhance Disaster Situational A...Mirjam-Mona
 
Owod 01-2-andrew young-corporatedata
Owod 01-2-andrew young-corporatedataOwod 01-2-andrew young-corporatedata
Owod 01-2-andrew young-corporatedataSafecast
 
Crisis Event Extraction Service (CREES) – Automatic Detection and Classificat...
Crisis Event Extraction Service (CREES) – Automatic Detection and Classificat...Crisis Event Extraction Service (CREES) – Automatic Detection and Classificat...
Crisis Event Extraction Service (CREES) – Automatic Detection and Classificat...Gregoire Burel
 
Renaissance Essay. Renaissance Essays. The Renaissance- in class essay assign...
Renaissance Essay. Renaissance Essays. The Renaissance- in class essay assign...Renaissance Essay. Renaissance Essays. The Renaissance- in class essay assign...
Renaissance Essay. Renaissance Essays. The Renaissance- in class essay assign...Maria Watson
 
Integrated Global Early Warning and Response System
Integrated Global Early Warning and Response SystemIntegrated Global Early Warning and Response System
Integrated Global Early Warning and Response SystemInSTEDD
 
Evolve: InSTEDD's Global Early Warning and Response System
Evolve: InSTEDD's Global Early Warning and Response SystemEvolve: InSTEDD's Global Early Warning and Response System
Evolve: InSTEDD's Global Early Warning and Response SystemTaha Kass-Hout, MD, MS
 
Mobile App for Disaster Management & Information Technology in Emergency Prep...
Mobile App for Disaster Management & Information Technology in Emergency Prep...Mobile App for Disaster Management & Information Technology in Emergency Prep...
Mobile App for Disaster Management & Information Technology in Emergency Prep...Associate Professor in VSB Coimbatore
 
Classifying Crises-Information Relevancy with Semantics
Classifying Crises-Information Relevancy with SemanticsClassifying Crises-Information Relevancy with Semantics
Classifying Crises-Information Relevancy with SemanticsCOMRADES project
 
Identifying and Characterizing User Communities on Twitter during Crisis Events
Identifying and Characterizing User Communities on Twitter during Crisis EventsIdentifying and Characterizing User Communities on Twitter during Crisis Events
Identifying and Characterizing User Communities on Twitter during Crisis EventsIIIT Hyderabad
 
Deep Learning Approach for Enhanced Cyber Threat Indicators in Twitter Stream
Deep Learning Approach for Enhanced Cyber Threat Indicators in Twitter StreamDeep Learning Approach for Enhanced Cyber Threat Indicators in Twitter Stream
Deep Learning Approach for Enhanced Cyber Threat Indicators in Twitter StreamSimranKetha
 
Processing Social Media Messages in Mass Emergency: A Survey
Processing Social Media Messages in Mass Emergency: A SurveyProcessing Social Media Messages in Mass Emergency: A Survey
Processing Social Media Messages in Mass Emergency: A SurveyMuhammad Imran
 
Machine Learning, Data Mining, and
Machine Learning, Data Mining, and Machine Learning, Data Mining, and
Machine Learning, Data Mining, and butest
 
Classification of Disastrous Tweets on Twitter using BERT Model
Classification of Disastrous Tweets on Twitter using BERT ModelClassification of Disastrous Tweets on Twitter using BERT Model
Classification of Disastrous Tweets on Twitter using BERT ModelIRJET Journal
 

Similar to PhaseVis: What, When, Where, and Who in Visualizing the Four Phases of Emergency Management Through the Lens of Social Media (20)

Rfs & social media
Rfs & social mediaRfs & social media
Rfs & social media
 
Crisis Information Processing - with the power of A.I.
Crisis Information Processing - with the power of A.I.Crisis Information Processing - with the power of A.I.
Crisis Information Processing - with the power of A.I.
 
Em2 0 Empa
Em2 0 EmpaEm2 0 Empa
Em2 0 Empa
 
05 june2013ezell uasi_nationalhls_2013
05 june2013ezell uasi_nationalhls_201305 june2013ezell uasi_nationalhls_2013
05 june2013ezell uasi_nationalhls_2013
 
Strong Angel - an Evolution in Preparedness
Strong Angel - an Evolution in PreparednessStrong Angel - an Evolution in Preparedness
Strong Angel - an Evolution in Preparedness
 
Helping Crisis Responders Find the Informative Needle in the Tweet Haystack
Helping Crisis Responders Find the Informative Needle in the Tweet HaystackHelping Crisis Responders Find the Informative Needle in the Tweet Haystack
Helping Crisis Responders Find the Informative Needle in the Tweet Haystack
 
Utilizing Community Volunteered Information to Enhance Disaster Situational A...
Utilizing Community Volunteered Information to Enhance Disaster Situational A...Utilizing Community Volunteered Information to Enhance Disaster Situational A...
Utilizing Community Volunteered Information to Enhance Disaster Situational A...
 
Owod 01-2-andrew young-corporatedata
Owod 01-2-andrew young-corporatedataOwod 01-2-andrew young-corporatedata
Owod 01-2-andrew young-corporatedata
 
Crisis Event Extraction Service (CREES) – Automatic Detection and Classificat...
Crisis Event Extraction Service (CREES) – Automatic Detection and Classificat...Crisis Event Extraction Service (CREES) – Automatic Detection and Classificat...
Crisis Event Extraction Service (CREES) – Automatic Detection and Classificat...
 
Renaissance Essay. Renaissance Essays. The Renaissance- in class essay assign...
Renaissance Essay. Renaissance Essays. The Renaissance- in class essay assign...Renaissance Essay. Renaissance Essays. The Renaissance- in class essay assign...
Renaissance Essay. Renaissance Essays. The Renaissance- in class essay assign...
 
Integrated Global Early Warning and Response System
Integrated Global Early Warning and Response SystemIntegrated Global Early Warning and Response System
Integrated Global Early Warning and Response System
 
Evolve: InSTEDD's Global Early Warning and Response System
Evolve: InSTEDD's Global Early Warning and Response SystemEvolve: InSTEDD's Global Early Warning and Response System
Evolve: InSTEDD's Global Early Warning and Response System
 
Mobile App for Disaster Management & Information Technology in Emergency Prep...
Mobile App for Disaster Management & Information Technology in Emergency Prep...Mobile App for Disaster Management & Information Technology in Emergency Prep...
Mobile App for Disaster Management & Information Technology in Emergency Prep...
 
Classifying Crises-Information Relevancy with Semantics
Classifying Crises-Information Relevancy with SemanticsClassifying Crises-Information Relevancy with Semantics
Classifying Crises-Information Relevancy with Semantics
 
Identifying and Characterizing User Communities on Twitter during Crisis Events
Identifying and Characterizing User Communities on Twitter during Crisis EventsIdentifying and Characterizing User Communities on Twitter during Crisis Events
Identifying and Characterizing User Communities on Twitter during Crisis Events
 
Akram.pptx
Akram.pptxAkram.pptx
Akram.pptx
 
Deep Learning Approach for Enhanced Cyber Threat Indicators in Twitter Stream
Deep Learning Approach for Enhanced Cyber Threat Indicators in Twitter StreamDeep Learning Approach for Enhanced Cyber Threat Indicators in Twitter Stream
Deep Learning Approach for Enhanced Cyber Threat Indicators in Twitter Stream
 
Processing Social Media Messages in Mass Emergency: A Survey
Processing Social Media Messages in Mass Emergency: A SurveyProcessing Social Media Messages in Mass Emergency: A Survey
Processing Social Media Messages in Mass Emergency: A Survey
 
Machine Learning, Data Mining, and
Machine Learning, Data Mining, and Machine Learning, Data Mining, and
Machine Learning, Data Mining, and
 
Classification of Disastrous Tweets on Twitter using BERT Model
Classification of Disastrous Tweets on Twitter using BERT ModelClassification of Disastrous Tweets on Twitter using BERT Model
Classification of Disastrous Tweets on Twitter using BERT Model
 

Recently uploaded

2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...Martijn de Jong
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?Antenna Manufacturer Coco
 
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEarley Information Science
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?Igalia
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processorsdebabhi2
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slidevu2urc
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessPixlogix Infotech
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreternaman860154
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking MenDelhi Call girls
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Servicegiselly40
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerThousandEyes
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)Gabriella Davis
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityPrincipled Technologies
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slidespraypatel2
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationSafe Software
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfEnterprise Knowledge
 

Recently uploaded (20)

2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?
 
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your Business
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreter
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Service
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slides
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
 

PhaseVis: What, When, Where, and Who in Visualizing the Four Phases of Emergency Management Through the Lens of Social Media

  • 1. ISCRAM 2013, May 12-15 1 PhaseVis: Visualizing the Four Phases of Emergency Management Through the Lens of Social Media Seungwon Yang et al. Department of Computer Science, Virginia Tech 5/13/2013
  • 2. Outline 1. Motivation 2. Hurricane Isaac 3. Approach (Selection, Classification, Visualization) 4. PhaseVis in Action 5. Limitations 6. Discussion ISCRAM 2013, May 12-15 2
  • 3. 1. Motivation  Four Phases of Emergency Management Model  FEMA training material adds ‘Prevention/Protection’ http://training.fema.gov/EMIWeb/IS/IS230B/IS230bCourse.pdf ISCRAM 2013, May 12-15 3 Response Recovery Mitigation Preparedne ss
  • 4. 2. Hurricane Isaac: Trajectory ISCRAM 2013, May 12-15 4 8/24 Cuba, Hispaniola: approx. 30 died 8/28-29 Mississippi River, Georgia, Port Fourchon, LA: 9 died After 8/30 N. Louisiana: degenerated to tropical depression *Image by Cyclonebiskit (Wikipedia) 8/21 Tropical storm Isaac 8/19-20 Extratropical cyclone
  • 5. ISCRAM 2013, May 12-15 5 Disaster Tweets with emergency orgs, agency names Visualiza on & Interac on Manual Labeling Training Data Trained classifica on model Cleaned Tweets Original Tweets Original & Retweets Classified Tweets Select and Preprocess Tweets Classify into 4 phases Implement visualization & interaction 3. Overall Approach
  • 6.  Tweet collection using ‘#isaac’ with yourTwapperKeeper  Situation report & Information sharing  Majority of tweets  Embedded URLs: news webpages, videos, photographs  Personal activity report  Very few ISCRAM 2013, May 12-15 6 3. Tweet Collection
  • 7.  Approx. 56,000 English tweets collected with ‘#Isaac’  5,677 tweets (10%) with reference to Red Cross, FEMA, or Salvation Army  1,453 non-retweets  1,121 manually labeled with one of four phases (response, recovery, mitigation, preparedness) ISCRAM 2013, May 12-15 7 3. Building a Dataset (1/2)
  • 8.  Tweet text + resource title ISCRAM 2013, May 12-15 8 Nice article abt our Dir. Of emerg srvcs @leopratte in #Louisiana organizing #redcross #Isaac relief http://t.co/D4RPr33n 3. Building a Dataset (2/2)
  • 9. ISCRAM 2013, May 12-15 9 Response More than 4,700 people in as many as 80 shelters in 7 states overnight; more than 3,000 #RedCross workers (37 from KC region) at #Isaac Recovery FEMA announces that federal aid has been made available for the state of Louisiana. #Isaac Mitigation FEMA mitigations advisers to offer rebuilding tips in St. Bernard and Ascension Parishes. http://t.co/ZziRGOGw #Isaac Preparednes s Very cool app! MT @redcross: Our hurricane app has info on #RedCross shelters, a toolkit w flashlight, alarm http://t.co/E7o1rtJK #Isaac 3. Examples of 4 Phases
  • 10.  SVM multiclass with linear kernel  Large num. of features, small num. of training examples  Naïve Bayes multinomial  Bag-of-words model fits well for tweet data  Random forest  One of the robust algorithms for text classification ISCRAM 2013, May 12-15 10 3. Classification Algorithms
  • 11.  TF, normalization, stemming applied  Tuned classifier, 10 fold cross-validation ISCRAM 2013, May 12-15 11 Precision Weighted F Measure Naïve Bayes multinomial 77.87% 0.782 Random forest 76.27% 0.754 SVM multiclass (linear kernel) 80.82% Reported slightly lower than Naïve Bayes multinomial 3. Classification Cross-Validation
  • 12. ISCRAM 2013, May 12-15 12 3. Tweet Visualization WHAT WHEN WHERE WHO
  • 13.  WHAT (Phases, List)  Phases: ThemeRiver, D3 visualization toolkit  Tweet List: JqGrid Library  WHEN (Timeline)  JavaScript  WHERE (user locations)  Google Maps API  WHO (user mention network)  Gephi graph format, Sigma.js ISCRAM 2013, May 12-15 13 3. Visualization Implementation
  • 14. ISCRAM 2013, May 12-15 14 4. PhaseVis in Action (8/23-8/24)
  • 15.  Majority of tweets in Preparedness phase (84%)  Content: fill up the gas tank, hurricane App, preparedness tips, replace food/water in emergency kit, etc…  Clustered around  Red Cross, FEMA, & CraigatFEMA  Study focus was rather on the US (English tweets)  Spanish tweets from Cuba, Hispaniola not considered  Unable to understand phases in such areas ISCRAM 2013, May 12-15 15 4. Summary (8/23-8/24)
  • 16. ISCRAM 2013, May 12-15 16 4. PhaseVis in Action (8/28-8/29) - Mainly in Louisiana, Mississippi, Georgia -
  • 17.  High increase in tweet volume  Isaac landed in the US in 8/28 with hurricane strength  Response (20%), Recovery (34%), Mitigation (0%), Preparedness (46%)  Content:  Recruiting volunteers (Response, Recovery)  Asking for donations/support (Recovery)  RT regarding ‘Mitt Romney’  Providing shelters (Response)… ISCRAM 2013, May 12-15 17 4. Tweet Details (8/28-8/29)
  • 18. ISCRAM 2013, May 12-15 18 4. PhaseVis in Action (9/5-9/7) - US continued -
  • 19.  Mostly Recovery phase (75%), followed by continued Response actions…  Lots of activities in New Orleans, Baton Rouge, Louisiana  Active tweet account: FEMA, Red Cross, RedCrossSELA (South East Louisiana) ISCRAM 2013, May 12-15 19 4. Tweet Details (9/5-9/7)
  • 20. ISCRAM 2013, May 12-15 20 5. Limitations  Language  Only English tweets considered  Unable to analyze Spanish tweets when Isaac hit Cuba & Hispaniola  Small data set  Only tweets containing FEMA, Red Cross & Salvation Army  E.g., RedCrossSELA, SalvationArmy, craigatFEMA, …  Approx. 10% of tweets had those names
  • 21. ISCRAM 2013, May 12-15 21 6. Discussion  What are other valuable information to uncover from disaster tweets and why are they important?  Sentiment, Reliability of tweets  Embedded URLs: news articles, images, videos…  ??  To what extent can tweet analysis actually help emergency managers in the field?  Identification of ‘actionable’ tweets from affected areas, victims, and witnesses…  ??
  • 22.  NSF for funding: IIS-0916733 (CTRnet project)  Internet Archive for collaboration  Big thanks to co-authors who couldn’t come here  Haeyong Chung, Xiao Lin, Sunshin Lee, Liangzhe Chen, Andy Wood, and the CTRnet Team ISCRAM 2013, May 12-15 22 Acknowledgment
  • 25. Evaluation  Preprocessing & Accuracy ISCRAM 2013, May 12-15 25 TF IDF Normali zation Naïve Bayes Multinomial SVM Multiclass 76% 80.1% X 77% 80.4% X 60% 78.8% X X 78.1% X 75% 80.4% X X 78% 80.8% X X 63% 78.9% X X X 79.0%
  • 26. ISCRAM 2013, May 12-15 26 3. Visualization: Phase View
  • 27. ISCRAM 2013, May 12-15 27 Overview Detail 3. Visualization: Social Network View
  • 28. ISCRAM 2013, May 12-15 28 3. Visualization: Location View
  • 29. ISCRAM 2013, May 12-15 29 Is_R (Retweet check) Tweet Text Phases Date 3. Visualization: Tweet View
  • 30. Use Case & Demo http://spare05.dlib.vt.edu/~ctrvis/phasevis/ind ex_may.html ISCRAM 2013, May 12-15 30
  • 31. ISCRAM 2013, May 12-15 31
  • 32. ISCRAM 2013, May 12-15 32

Editor's Notes

  1. 8/28 morning – reached hurricane strength
  2. Goal: finding four phases in disaster tweets
  3. (QUESTION for Audience)Often ‘NULL’ title if attempts to access URLs after a month.Sometimes, title is almost the same as tweet contentAlso note the informal word usage: ‘abt’, ‘emerg’, ‘srvcs’