Analyzing and Visualizing Disaster Phases from Social Media

Analyzing and Visualizing Disaster Phases from Social Media

Analyzing and Visualizing Disaster Phases from Social Media Streams Group VizDisasters: Liangzhe Chen, Xaio Lin, Andrew Wood Client: Seungwon Yang Information Storage & Retrieval Final Presentation 12/4/2012 Virginia Tech Motivation CTRnet: archiving disaster-related online data in collaboration with the Internet Archive Tweets during disasters: quick alternative to

cell phones Large dataset to pull from for researchers & responders Four Phases of Emergency Management Four phases: Response Recovery Mitigation Preparedness Professional and personal activities Four Phases in Tweets Reporting situation / sharing information Majority

For hurricane: rain, flood, wind, cloud, weather forecast Photographs (Instagram) Reporting personal activities Very few 02/11/2020 ProjVisDisaster 4 Four Phases in Tweets Reporting professional activities 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 Preparedness Very cool app! MT @redcross: Our hurricane app has info on #RedCross shelters, a toolkit w flashlight, alarm http://t.co/E7o1rtJK #Isaac 02/11/2020

ProjVisDisaster 5 Our Approach 02/11/2020 ProjVisDisaster 6 Our Approach Machine learning Extract professional activities Classify professional activities into four phases

Visualization Phase view, tweet view, social network view, map view Use case / Demo 02/11/2020 ProjVisDisaster 7 Learning Professional Activities in Four Phases

02/11/2020 ProjVisDisaster 8 Learning Professional Activities in Four Phases Preprocessing Building dataset Vectorization Classification

Algorithms Evaluation 02/11/2020 ProjVisDisaster 9 Building dataset Focus on tweets about professional activities Based on keywords of known organizations FEMA Red Cross (RedCross) Salvation Army (SalvationArmy)

02/11/2020 ProjVisDisaster 10 Building dataset Combining tweet and resource title Mitigation specialists are offering free rebuilding tips in five parishes. http://t.co/hwXajm6X #Isaac 02/11/2020 ProjVisDisaster 11

Building dataset Overview of Issac dataset About 56,000 English tweets during hurricane Issac 5,677 tweets with reference to FEMA, Red Cross or Salvation Army 1,453 without re-tweets 1,121 manually labeled explicitly with one of the four phases, response, recovery, mitigation or preparedness 02/11/2020 ProjVisDisaster 12

Vectorization tf transform idf transform Normalization Stemming (Porter stemmer) 02/11/2020 ProjVisDisaster

13 Algorithms Nave Bayes Nave Bayes Multinomial Random Forest SVM Multiclass 02/11/2020 ProjVisDisaster

14 Evaluation Tuned classifier, 10 fold cross-validation Accuracy Weighted F Measure Nave Bayes 70.47% 0.723 Nave Bayes

Multinomial 77.87% 0.782 Random Forest 76.27% 0.754 SVM Multiclass 80.82%

Reported slightly lower than nave bayes multinomial 02/11/2020 ProjVisDisaster 15 Evaluation Preprocessing v.s. Accuracy TF IDF

Normaliz Nave Bayes ation Multinomial X X X X X X X X

X 02/11/2020 76% 80.1% 77% 80.4% 60% 78.8%

X X SVM Multiclass 78.1% 75% 80.4% 78% 80.8%

63% 78.9% X 79.0% ProjVisDisaster 16 Visualizing Four Phases 02/11/2020 ProjVisDisaster

17 Visualizing Four Phases Phase view ThemeRiver, D3 library Tweet view JqGrid Library Social Network View Gephi Map View Google Geocoding API 02/11/2020

ProjVisDisaster 18 Phase view 02/11/2020 ProjVisDisaster 19 Tweet view 02/11/2020

ProjVisDisaster 20 Social Network View 02/11/2020 ProjVisDisaster 21 Map view 02/11/2020

ProjVisDisaster 22 Use Case & Demo http://spare05.dlib.vt.edu/~ctrvis/phasevis/ 02/11/2020 ProjVisDisaster 23 Use Case

02/11/2020 ProjVisDisaster 24 Use Case 02/11/2020 ProjVisDisaster 25 Summary and Future Work Summary:

Analysis/classification of disaster tweets into phases Multi-view visualization Future challenges: Automated professional organization extraction Processing of personal tweets Application to other disasters 02/11/2020 ProjVisDisaster 26 Acknowledgements Haeyong Chung

Sunshin Lee 02/11/2020 ProjVisDisaster 27

Recently Viewed Presentations

  • Part 2 - Rotary District 5220 - District 5220

    Part 2 - Rotary District 5220 - District 5220

    The Rotary Club of Metro sponsored a scholar to study disease prevention at a two-year master's degree program in another country. Before departure, the scholar attended a district orientation to review expectations, sexual harassment prevention, Rotary history, and financial and...
  • In the name of ALLAH the beneficent the merciful

    In the name of ALLAH the beneficent the merciful

    IMADJUST. The intensity of digital image of yarn is adjusted in order to evident the minute. details more clearly. Image cropping and pattern recognition. Phenomenon of alternating color intensities visualized in gray scale image of yarn present in yarn texture...
  • Tiny House - fcs246 Visual Communication for Interior Design

    Tiny House - fcs246 Visual Communication for Interior Design

    -seating area use a drop down table to create an eating area-Be sure to add wheels or felt to the legs of. furniture for mobility
  • The Five Paragraph Order - Weebly

    The Five Paragraph Order - Weebly

    An order should contain everything that a commander cannot do himself, but nothing else."--Count Helmut Von-Moltke At OCS, you will use a version of the five-paragraph operations order format specially adapted for use by companies, platoons, and squads. Five Paragraph...
  • Stars

    Stars

    The temperature and color of a star depends directly on its size. Bigger=hotter, more gravity makes it burn faster raising its temperature. Larger stars have a shorter life than smaller stars due to the faster burning of hydrogen fuel.
  • Code Injection Attacks on HTML5-based Mobile Apps ...

    Code Injection Attacks on HTML5-based Mobile Apps ...

    Code Injection Attacks on HTML5-based Mobile Apps: Characterization, Detection and Mitigation. Xing Jin, Xunchao Hu, Kailiang Ying, Wenliang Du, Heng Yin and Gautam Nagesh Peri
  • Charles Dickens & Victorian england

    Charles Dickens & Victorian england

    Charles Dickens. 1812-1870. Writer and social critic. Was a child worker at age 12 in a factory when his family went to jail for debt- he HATED it. Studied law when his family left jail
  • Lesson 17 David Plans a Temple 2 Samuel

    Lesson 17 David Plans a Temple 2 Samuel

    4 - top two Freddy. 5 - bottom two Randall. ... Ac 2:25 "For David says concerning Him: 'I foresaw the LORD always before my face, For He is at my right hand, that I may not be shaken. 26...