東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Exploration of Public Sentiment as a...

Luna, Sergio.

Linked to FindBook

Google Book

Amazon

博客來

Exploration of Public Sentiment as an Indicator of Public Response to Natural Disasters: An Analysis of Hurricane Scenarios.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Exploration of Public Sentiment as an Indicator of Public Response to Natural Disasters: An Analysis of Hurricane Scenarios./
Author:	Luna, Sergio.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2019,
Description:	180 p.
Notes:	Source: Dissertations Abstracts International, Volume: 80-12, Section: B.
Contained By:	Dissertations Abstracts International80-12B.
Subject:	Civil engineering. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=13882484
ISBN:	9781392240762

Exploration of Public Sentiment as an Indicator of Public Response to Natural Disasters: An Analysis of Hurricane Scenarios.
Luna, Sergio.

Exploration of Public Sentiment as an Indicator of Public Response to Natural Disasters: An Analysis of Hurricane Scenarios. - Ann Arbor : ProQuest Dissertations & Theses, 2019 - 180 p.

Source: Dissertations Abstracts International, Volume: 80-12, Section: B.

Thesis (Ph.D.)--Stevens Institute of Technology, 2019.

This item must not be added to any third party search indexes.

The adoption of social media applications offers new alternatives for individuals to interact during emergency situations, expanding from a classical face-to-face communication to a virtual environment where a vast number of individuals can be reached simultaneously. The implementation of this technology results in a vast amount of data being generated, disseminated and considered in the individual's decision-making process. Despite expert recognition that the integration of social media into disaster response strategies may contribute to situational awareness, they have been reluctant to fully adopt it because it has the potential to exacerbate the negative impact of disasters.To understand why, this dissertation contributes to the emergency management domain by exploring how social media can be leveraged during natural disasters. Specifically, the research contributes from three different angles: 1) Investigating the body of knowledge to understand social media in the context of disaster response, 2) Understanding the implications of bias when dealing with social media extracted data, along with its repercussions when selecting leading indicators for evacuation, and 3) Exploring how public sentiment emerges and is shaped by social interactions. As each of these angles addresses a different aspect of social media impacts, three different studies were conducted. First, a grounded theory approach was implemented to identify the characteristics of social media that could be exploited, as well as those that inhibit adoption. The analysis identified three major benefits including: 1) support situational awareness, 2) faster information diffusion, and 3) monitoring activities and coordination. Factors that inhibit the adoption were clustered into social and technical limitations. Second, the analysis of public sentiment and the implications of bias towards three of the costliest hurricanes on US record, hurricane Harvey, Irma and Maria, was conducted considering Twitter and U.S official environmental sensors data. The effects of bias were analyzed by comparing the outcomes of two well-known methods for model selection, Lasso method with Bootstrapping and a bias correction causal inference model. Our findings indicate significant differences in predictions between the traditional and bias correction model, suggesting emergency managers to correct for biases when seeking to integrate public sentiment in their strategies. Third, a Brownian agent-based model of collective emotions, with foundations on the psychological theory of Circumplex Model of Affect, was developed to explore how public's collective sentiment is altered according to the available communication methods and the type of community the individual is part of. The analyzed communication preferences include face-to-face and social media, while the types of communities explored are tight-knit, traditional, urban modern and social media. Insights from the agent-based model suggest community structure highly influences the rate and shape of public sentiment. Also, social media's faster dissemination rates play a significant role on the resulting public sentiment. Emergency managers benefit from the developed simulation as they can have a reference map for how a community's communication preference influences the resulting public sentiment.Overall, the findings of this research will enable emergency managers to better understand the role of social media during disasters as well as inform the development of emergency management strategies by recognizing how social media can impact the dynamics of public perception and behavior.

ISBN: 9781392240762Subjects--Topical Terms:

860360
Civil engineering.
Subjects--Index Terms:

Agent-based models

Exploration of Public Sentiment as an Indicator of Public Response to Natural Disasters: An Analysis of Hurricane Scenarios.
LDR:05050nmm a2200409 4500 001 2274408
005 20201202125013.5
008 220629s2019 ||||||||||||||||| ||eng d
020 $a 9781392240762
035 $a (MiAaPQ)AAI13882484
035 $a (MiAaPQ)stevens:10611
035 $a AAI13882484
040 $a MiAaPQ $c MiAaPQ
100 1 $a Luna, Sergio. $3 3551897
245 1 0 $a Exploration of Public Sentiment as an Indicator of Public Response to Natural Disasters: An Analysis of Hurricane Scenarios.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2019
300 $a 180 p.
500 $a Source: Dissertations Abstracts International, Volume: 80-12, Section: B.
500 $a Publisher info.: Dissertation/Thesis.
500 $a Advisor: Pennock, Michael J.
502 $a Thesis (Ph.D.)--Stevens Institute of Technology, 2019.
506 $a This item must not be added to any third party search indexes.
506 $a This item must not be sold to any third party vendors.
520 $a The adoption of social media applications offers new alternatives for individuals to interact during emergency situations, expanding from a classical face-to-face communication to a virtual environment where a vast number of individuals can be reached simultaneously. The implementation of this technology results in a vast amount of data being generated, disseminated and considered in the individual's decision-making process. Despite expert recognition that the integration of social media into disaster response strategies may contribute to situational awareness, they have been reluctant to fully adopt it because it has the potential to exacerbate the negative impact of disasters.To understand why, this dissertation contributes to the emergency management domain by exploring how social media can be leveraged during natural disasters. Specifically, the research contributes from three different angles: 1) Investigating the body of knowledge to understand social media in the context of disaster response, 2) Understanding the implications of bias when dealing with social media extracted data, along with its repercussions when selecting leading indicators for evacuation, and 3) Exploring how public sentiment emerges and is shaped by social interactions. As each of these angles addresses a different aspect of social media impacts, three different studies were conducted. First, a grounded theory approach was implemented to identify the characteristics of social media that could be exploited, as well as those that inhibit adoption. The analysis identified three major benefits including: 1) support situational awareness, 2) faster information diffusion, and 3) monitoring activities and coordination. Factors that inhibit the adoption were clustered into social and technical limitations. Second, the analysis of public sentiment and the implications of bias towards three of the costliest hurricanes on US record, hurricane Harvey, Irma and Maria, was conducted considering Twitter and U.S official environmental sensors data. The effects of bias were analyzed by comparing the outcomes of two well-known methods for model selection, Lasso method with Bootstrapping and a bias correction causal inference model. Our findings indicate significant differences in predictions between the traditional and bias correction model, suggesting emergency managers to correct for biases when seeking to integrate public sentiment in their strategies. Third, a Brownian agent-based model of collective emotions, with foundations on the psychological theory of Circumplex Model of Affect, was developed to explore how public's collective sentiment is altered according to the available communication methods and the type of community the individual is part of. The analyzed communication preferences include face-to-face and social media, while the types of communities explored are tight-knit, traditional, urban modern and social media. Insights from the agent-based model suggest community structure highly influences the rate and shape of public sentiment. Also, social media's faster dissemination rates play a significant role on the resulting public sentiment. Emergency managers benefit from the developed simulation as they can have a reference map for how a community's communication preference influences the resulting public sentiment.Overall, the findings of this research will enable emergency managers to better understand the role of social media during disasters as well as inform the development of emergency management strategies by recognizing how social media can impact the dynamics of public perception and behavior.
590 $a School code: 0733.
650 4 $a Civil engineering. $3 860360
650 4 $a Social structure. $3 528995
653 $a Agent-based models
653 $a Bias correction
653 $a Causal inference
653 $a Emergency management
653 $a Network science
653 $a Social medua
690 $a 0543
690 $a 0700
710 2 $a Stevens Institute of Technology. $b Systems Engineering. $3 3186847
773 0 $t Dissertations Abstracts International $g 80-12B.
790 $a 0733
791 $a Ph.D.
792 $a 2019
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=13882484