Towards Natural Disasters Detection from Twitter Using Topic Modelling

Hagras, Mohamed; Hassan, Ghada; Farag, Nadine

doi:10.1109/eecs.2017.57

Cited by 11 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By applying topic modeling, researchers can systematically organize and classify the diverse range of topics discussed in social media posts. Researchers have used these techniques to automatically process and categorize textual data into relevant categories, such as types of disasters, affected regions, and key events [15,[32][33][34]. These techniques enable efficient and automated classification of large volumes of text, facilitating the identification of critical information for emergency management and response.…”

Section: Literature Reviewmentioning

confidence: 99%

Unveiling Key Themes and Establishing a Hierarchical Taxonomy of Disaster-Related Tweets: A Text Mining Approach for Enhanced Emergency Management Planning

2023

View full text Add to dashboard Cite

Effectively harnessing the power of social media data for disaster management requires sophisticated analysis methods and frameworks. This research focuses on understanding the contextual information present in social media posts during disasters and developing a taxonomy to effectively categorize and classify the diverse range of topics discussed. First, the existing literature on social media analysis in disaster management is explored, highlighting the limitations and gaps in current methodologies. Second, a dataset comprising real-time social media posts related to various disasters is collected and preprocessed to ensure data quality and reliability. Third, three well-established topic modeling techniques, namely Latent Dirichlet Allocation (LDA), Latent Semantic Analysis (LSA), and Non-Negative Matrix Factorization (NMF), are employed to extract and analyze the latent topics and themes present in the social media data. The contributions of this research lie in the development of a taxonomy that effectively categorizes and classifies disaster-related social media data, the identification of key latent topics and themes, and the extraction of valuable insights to support and enhance emergency management efforts. Overall, the findings of this research have the potential to transform the way emergency management and response are conducted by harnessing the power of social media data. By incorporating these insights into decision-making processes, emergency managers can make more informed and strategic choices, resulting in more efficient and effective emergency response strategies. This, in turn, leads to improved outcomes, better utilization of resources, and ultimately, the ability to save lives and mitigate the impacts of disasters.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Unveiling Key Themes and Establishing a Hierarchical Taxonomy of Disaster-Related Tweets: A Text Mining Approach for Enhanced Emergency Management Planning

2023

View full text Add to dashboard Cite

show abstract

“…Topic Modeling does not require a labelling process in documents [12]. Topic Modeling to identify a set of words from large data to generate topics based on the probability value distribution of each word in a document [13].…”

Section: Topic Modelingmentioning

confidence: 99%

Topic Modeling for Support Ticket using Latent Dirichlet Allocation

Wiranto

Uswatunnisa

2022

J. RESTI (Rekayasa Sist. Teknol. Inf.)

View full text Add to dashboard Cite

In the business world, communication over customers must be built properly to make it easier for companies to find out what customers want. Support ticket is one of the business instrument for communication between the customers and the companies. Through a support ticket, customers can respond, complain or ask questions about products with a support team. Increasing the business process of the companies will be increasing the support ticket volume that should be handled by support team. It also has a value for analysis to get business intelligence decision. With that chance, an efficient data processing method is needed to find topics are being discussed by customers. One way that can be used to solve this problem is Topic Modeling. This research uses several parameters the number of topics, alpha value, beta value, iteration, and random seed. With this combination of parameters, the best results based on evaluation of human judgement and topic coherence with 5 topics, an alpha value of 50, a beta value of 0.01, 100 iterations, and 50 random seeds. The five topics interpretation consists of hosting migration, error problems in wordpress, domain email settings and domain transfer, ticketing and transaction processing. The total of 5 topics has a coherence value of 0.507897.

show abstract

“…According to the kind of event, the method of detection, and the intended use, (Table I) shows a taxonomy of research on Twitter event detection based on event type and approach. [27] Open semi-Supervised General Event Detection Hagras et al (2017) [21] classified and evaluated tweets related to the Japan Tsunami using the Latent Dircherilet Allocation (LDA) topic analysis approach. Selected 196 tweets for the test set and 6700 tweets for the training set, resulting in 76% accuracy with successful detection.…”

Section: A Event Detection On Twittermentioning

confidence: 99%

Arabic Location Named Entity Recognition for Tweets using a Deep Learning Approach

Alzaidi¹,

Abushark²,

Khan³

2022

IJACSA

View full text Add to dashboard Cite

Social media sites like Twitter have emerged in recent years as a major data source utilized in a variety of disciplines, including economics, politics, and scientific study. To extract pertinent data for decision-making and behavioral analysis, one can use Twitter data. To extract event location names and entities from colloquial Arabic texts using deep learning techniques, this study proposed Named Entity Recognition (NER) and Linking (NEL) models. Google Maps was also used to obtain up-to-date details for each extracted site and link them to the geographical coordination. Our method was able to predict 40% and 48% of the locations of tweets at the regional and city levels, respectively, while the F-measure was able to reliably identify and detect 63% of the locations of tweets at a single Point of Interest.

show abstract

Towards Natural Disasters Detection from Twitter Using Topic Modelling

Cited by 11 publications

References 14 publications

Unveiling Key Themes and Establishing a Hierarchical Taxonomy of Disaster-Related Tweets: A Text Mining Approach for Enhanced Emergency Management Planning

Unveiling Key Themes and Establishing a Hierarchical Taxonomy of Disaster-Related Tweets: A Text Mining Approach for Enhanced Emergency Management Planning

Topic Modeling for Support Ticket using Latent Dirichlet Allocation

Arabic Location Named Entity Recognition for Tweets using a Deep Learning Approach

Contact Info

Product

Resources

About