Harnessing the power of machine learning: Can Twitter data be useful in guiding resource allocation decisions during a natural disaster?

Reynard, Darcy; Shirgaokar, Manish

doi:10.1016/j.trd.2019.03.002

Cited by 62 publications

(31 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Thus, people are more likely to post a longer statement on Twitter to express their positive feelings about the reopening of the economy with useful information. A study analyzed Twitter data to understand sentiments of the citizens for allocating resources during Hurricane Irma in 2017 in Florida [41]. Upon analyzing data the study found that longer tweets are more likely to have useful information with sentiment contents.…”

Section: Discussion Of Resultsmentioning

confidence: 99%

“…The data is free, publicly available, and considered as an important source of information for researchers from different disciplines. Many previous studies collected information from ACS and leveraged with Twitter data to analyze sentiment of the people in the arena of public health [28], urban spaces [38], politics [39,40], disasters management [41], racial conflicts [30,42], and gender disparity [43]. Thus, linking Twitter data with Coronavirus data is a common practice among the researchers to evaluate the impacts of socioeconomic and demographic characteristics on the sentiments of the people towards a subject of interest.…”

Section: Us Census Data and Socioeconomic Analysismentioning

confidence: 99%

See 1 more Smart Citation

Twitter and Census Data Analytics to Explore Socioeconomic Factors for Post-COVID-19 Reopening Sentiment

Rahman

Ali

et al. 2020

SSRN Journal

View full text Add to dashboard Cite

Investigating and classifying sentiments of social media users (e.g., positive, negative) towards an item, situation, and system are very popular among the researchers. However, they rarely discuss the underlying socioeconomic factor associations for such sentiments. This study attempts to explore the factors associated with positive and negative sentiments of the people about reopening the economy, in the United States (US) amidst the COVID-19 global crisis. It takes into consideration the situational uncertainties (i.e., changes in work and travel pattern due to lockdown policies), economic downturn and associated trauma, and emotional factors such as depression. To understand the sentiment of the people about the reopening economy, Twitter data was collected, representing the 51 states including Washington DC of the US. Statewide socioeconomic characteristics of the people (e.g., education, income, family size, and employment status), built environment data (e.g., population density), and the number of COVID-19 related cases were collected and integrated with Twitter data to perform the analysis. A binary logit model was used to identify the factors that influence people toward a positive or negative sentiment. The results from the logit model demonstrate that family households, people with low education levels, people in the labor force, low-income people, and people with higher house rent are more interested in reopening the economy. In contrast, households with a high number of members and high income are less interested to reopen the economy. The accuracy of the model is good (i.e., the model can correctly classify 56.18% of the sentiments). The Pearson chi2 test indicates that overall this model has high goodness-of-fit. This study provides a clear indication to the policymakers where to allocate resources and what policy options they can undertake to improve the socioeconomic situations of the people and mitigate the impacts of pandemics in the current situation and as well as in the future.

show abstract

Section: Discussion Of Resultsmentioning

confidence: 99%

Section: Us Census Data and Socioeconomic Analysismentioning

confidence: 99%

Twitter and Census Data Analytics to Explore Socioeconomic Factors for Post-COVID-19 Reopening Sentiment

Rahman

Ali

et al. 2020

SSRN Journal

View full text Add to dashboard Cite

show abstract

“…ML has been framed as an effective approach for filtering out noisy and irrelevant information that accommodates the volume and speed of real‐time social media data during a crisis (Nguyen et al., 2016; Rao, Plotnick, & Hiltz, 2017). Previous research efforts have developed ML models to classify unstructured text content of messages generated on social media during crisis events (e.g., Buscaldi & Hernandez‐Farias, 2015; Reynard & Shirgaokar, 2019). For example, ML has been used to classify text content according to the stages of a crisis (Verma et al., 2011) and the sentiment of the public reacting to a crisis (Beigi, Hu, Maciejewski, & Liu, 2016).…”

Section: Related Workmentioning

confidence: 99%

“…Researchers also have built ML models to map content based on location markers in the text (Cresci, Cimino, Dell’Orletta, & Tesconi, 2015; Ghahremanlou, Sherchan, & Thom, 2015). Reynard and Shirgaokar (2019) used ML to assist with geolocation and sentiment classification of tweets to help guide resource allocation during a natural disaster. Nguyen, Alam, Ofli, and Imran (2017) used a type of DL called a convolution neural network (CNN) to attempt to automatically detect the level of crisis damage from images.…”

Section: Related Workmentioning

confidence: 99%

Using deep learning and social network analysis to understand and manage extreme flooding

Romascanu

Ker

Sieber

et al. 2020

Contingencies & Crisis Mgmt

View full text Add to dashboard Cite

Combining machine learning with social network analysis (SNA) can leverage vast amounts of social media data to better respond to crises. We present a case study using Twitter data from the March 2019 Nebraska floods in the United States, which caused over $1 billion in damage in the state and widespread evacuations of residents. We use a subset of machine learning, deep learning (DL), to classify text content of 11,982 tweets, and we integrate that with SNA to understand the structure of tweet interactions. Our DL approach pre‐trains our model with a DL language technique, BERT, and then trains the model using the standard training dataset to sort a dataset of tweets into classes tailored to crisis events. Several performance measures demonstrate that our two‐tiered trained model improves domain adaptation and generalization across different extreme weather event types. This approach identifies the role of Twitter during the damage containment stage of the flood. Our SNA identifies accounts that function as primary sources of information on Twitter. Together, these two approaches help crisis managers filter large volumes of data and overcome challenges faced by simple statistical models and other computational techniques to provide useful information during crises like flooding.

show abstract

“…These state-of-the-art platforms, which give us the geospatial information of the posts, are called Location Based Social Network (LBSN). This is one of the primary reasons to perform sentiment analysis on a LBSN like twitter for disaster management as it will give us the specific location of the disaster event [3]. LBSN includes any social media network, which handles geospatial data like twitter, facebook, Snapchat, Instagram and other similar platforms.…”

Section: Introductionmentioning

confidence: 99%

Sentiment Severity on Location-Based Social Network (LBSN) Data of Natural disasters

Shyamala*,

Kannan,

Dass. D

2020

IJRTE

View full text Add to dashboard Cite

Social media emerged as one of the key components to reach disaster affected people, as they supplement planning and operational coordination. Sentiment analysis was expended to identify, extract or characterize subjective information, such as opinions, expressed in a tweet. The sentiment expressed is analyzed and is classified as positive or negative sentiment, which is not versatile enough to capture the exact sentiment conveyed by the user. Opinion mining is a machine learning process used to extract information conveyed by the user in the form of text. In this paper, the lexical analysis to sentiment analysis of twitter data is employed. Conventionally, the sentiment is conveyed using the polarity of the data but in this paper, sentiment intensity is employed to convey the sentiments. Performing sentiment analysis on tweets gives us the sentiment intensity conveyed by the user, which in turn is used to calculate the severity of the disaster event specified by the user. Further, it is also used to classify the tweets based on their severity. This paper proposes a methodology to extract relevant sentiment information from Location Based Social Network (LBSN) and suggests a unique scale to classify this information to help disaster management authority.

show abstract

Harnessing the power of machine learning: Can Twitter data be useful in guiding resource allocation decisions during a natural disaster?

Cited by 62 publications

References 24 publications

Twitter and Census Data Analytics to Explore Socioeconomic Factors for Post-COVID-19 Reopening Sentiment

Twitter and Census Data Analytics to Explore Socioeconomic Factors for Post-COVID-19 Reopening Sentiment

Using deep learning and social network analysis to understand and manage extreme flooding

Sentiment Severity on Location-Based Social Network (LBSN) Data of Natural disasters

Contact Info

Product

Resources

About