Figure - uploaded by Sreenivasulu Madichetty
Content may be subject to copyright.
Source publication
Twitter is an excellent resource for communicating between the victims and organizations during a disaster. People share opinions, sympathies, situational information, etc., in the form of tweets during a disaster. Detecting the situational tweets is a challenging task, which is very helpful to both humanitarian organizations and victims. There is...
Context in source publication
Context 1
... and then apply the classifier with use of the low-level lexical and syntactic features for classifying the situational and non-situational tweets. The examples of fragment tweets are shown in Table 2. However, the authors mentioned that their method doesn't perform well for a mixture of situational and non-situational information tweets compared to the situational tweets (fragment) only. ...Citations
... There is a likelihood that a tweet will include both situational and non-situational information. In this research [40] attempts to solve the challenge of detecting the situational tweets using deep learning algorithm. In this study [41] compare the performance of deep neural networks to that of conventional classifiers to better understand the usefulness of deep neural networks This study [42] examines the use of social media in crisis management with a focus on volunteer groups and humanitarian organization. ...
All around the world, natural disasters have an impact on both human and animal life. In addition to causing major harm to property, animal life, etc. Natural disasters (including landslides, cloudbursts, heat waves, hurricanes, tsunamis, floods, tsunamis, earthquakes, and wildfires) affect thousands of lives each year throughout the world. Twitter, a social networking platform, users can share news, opinions, and personal stories. Due to the widespread availability of real-time data, numerous service agencies regularly analyze this data to spot crises, lower risk, and save lives. Humans, however, are unable to manually filter through the vast number of records and spot hazards in real-time to achieve this, it has been suggested in numerous studies to provide words in forms that computers can understand and on the word representations, use machine learning techniques to determine the meaning behind a post with accuracy. The community can monitor disasters by reporting hazards that are related to disaster occurrences on social media, which has been essential to emergency preparedness. This study examines how social media platform Twitter might be used in disaster-related research. It focuses on the most recent machine learning, deep learning, and disaster prediction techniques. Acquiring a thorough grasp of the numerous data kinds and their sources in relation to a variety of tasks and crisis management scenarios is another goal of the work. Additionally, The study also aims to offer a comprehensive analysis of the various data mining methods utilized for tackling various issues related to natural disasters as well as comprehensive directions on how to categorize tweets as "Related to Catastrophe" or "Not related with Catastrophe" using natural processing methods.
... The research conducted by Madichetty and Muthukumarasamy tackles the task of detecting situational tweets using various deep learning architectures, including Convolutional Neural Network (CNN), long short-term memory (LSTM), bi-directional long short-term memory (BLSTM), and bi-directional long short-term memory with attention (BLSTM attention) [67]. Notably, these deep learning models are applied to both English and Hindi language tweets to identify situational information during disasters. ...
... Some topics of ASTS require greater attention in future research endeavors: I Extractive-based ASTS techniques encompass a range of methodologies, including machine learning and optimization-based, statistical-based, graph-based, and fuzzylogic-based approaches. Within the machine learning realm, both supervised and unsupervised techniques aim to enhance accuracy [67] quality [52,58,63,65], and similarity [42,59]. Yet, a comprehensive evaluation often requires a broader focus, encompassing diversity, relevancy, completeness, readability, and redundancy. ...
... Unsupervised methods, while promising, encounter the complexities inherent in the diversity and noise of tweets. Optimizationbased techniques seek to optimize redundancy [44,54,66], diversity, coverage [67], and other objectives. However, their potential to enhance summary quality, accuracy, similarity, sparsity, and computational efficiency deserves deeper exploration. ...
The rapid expansion of social media platforms has resulted in an unprecedented surge of short text content being generated on a daily basis. Extracting valuable insights and patterns from this vast volume of textual data necessitates specialized techniques that can effectively condense information while preserving its core essence. In response to this challenge, automatic short text summarization (ASTS) techniques have emerged as a compelling solution, gaining significant importance in their development. This paper delves into the domain of summarizing short text on social media, exploring various types of short text and the associated challenges they present. It also investigates the approaches employed to generate concise and meaningful summaries. By providing a survey of the latest methods and potential avenues for future research, this paper contributes to the advancement of ASTS in the ever-evolving landscape of social media communication.
... In most of the literature on tweet classification, the datasets are initially created with manual annotations [8] so that the annotators can preprocess the individual tweets. Late the following recent trends, many researchers focused on machine learning based [3][4][5], deep learning based [11] approaches. In the recent past, most researchers have focused on transfer learning approaches such as voting [31,39], stacking [10,42], and BERT [33] to improve classification tasks. ...
... In related works, the authors [11] focus on the identification of tweets during disasters in English and Hindi. They compared the performance of various word embeddings and found that Crisis word embeddings performed the best. ...
... This operation is shown in equation [10] Let F be the set of filters, each having a size of "h" words (h < n). The output of the j-th filter (also known as a feature map) is given by equation [11] _ = [ _1 , _2 , . . . , _( − ℎ + 1) ] ...
In recent years, disaster tweet classification has garnered significant attention in natural language processing (NLP) due to its potential to aid disaster response and emergency management. The goal of disaster tweet classification is to automate the identification of informative tweets containing information related to various types of disasters, such as floods, earthquakes, wildfires, and more. This classification task plays a crucial role in real-time monitoring, situational awareness, and timely response coordination during emergency situations. In this context, we propose a deep parallel hybrid fusion model (DPHFM) that combines features extracted from Convolutional Neural Networks (CNNs) and Bidirectional Long Short-Term Memory (Bi-LSTM) as base learners. The extracted features from the base learners are combined using a fusion mechanism, and the resulting features are then reconstructed and supplied to a meta-learner as input for making predictions. The DPHFM is trained on disaster datasets, such as crisisMMD, which consists of seven natural disaster events. The model was thoroughly evaluated using various metrics, demonstrating an average performance improvement of 90–96%. Furthermore, the proposed model's performance surpassed that of other state-of-the-art models, showcasing its potential for disaster tweet classification using deep learning techniques.
... For example, existing content-based disaster summarization approaches propose different mechanisms to select the tweets into the summary, such as the presence of important words in the tweet [18], [54] and coverage of relevant concepts by the tweets. While the presence of important words is measured by the frequency and information content of the words [18], [54], relevant concepts are determined by segregating tweets into either relevant or irrelevant by semi-supervised learning [55] or supervised learning [19], [8], [56], [57]. Additionally, deep learning-based approaches have proposed different neural network architectures, such as the disaster-specific BERT model [52] or graph convolutional neural network-based model [51], to identify the tweets to be selected in summary. ...
The huge popularity of social media platforms, such as Twitter, attracts a large fraction of users to share real-time information and short situational messages during disasters. A summary of these tweets is required by the government organizations, agencies, and volunteers for efficient and quick disaster response. However, the huge influx of tweets makes it difficult to manually get a precise overview of ongoing events. To handle this challenge, several tweet summarization approaches have been proposed. In most of the existing literature, tweet summarization is broken into a two-step process where, in the first step, it categorizes tweets, and in the second step, it chooses representative tweets from each category. There are both supervised and unsupervised approaches found in the literature to solve the problem of first step. Supervised approaches require a huge amount of labeled data, which incurs cost as well as time. On the other hand, unsupervised approaches could not cluster tweet properly due to the overlapping keywords, vocabulary size, lack of understanding of semantic meaning, and so on, while, for the second step of summarization, existing approaches applied different ranking methods where those ranking methods are very generic, which fail to compute proper importance of a tweet with respect to a disaster. Both problems can be handled far better with proper domain knowledge. In this article, we exploited already existing domain knowledge by the means of ontology in both steps and proposed a novel disaster summarization method OntoDSumm. We evaluate this proposed method with six state-of-the-art methods using
$12$
disaster datasets. Evaluation results reveal that OntoDSumm outperforms the existing methods by approximately 2%–66% in terms of ROUGE-1 F1-score.
... Tweets were also used to extract the damages of disasters by applying various machine-learning methods, such as support vector machine, decision tree, and logistic regression [15]. Another study compared different neural networks built on top of various word embedding models of Tweets in Hindi on detecting situational information during various disasters [16]. However, to the best of our knowledge, no study has tried to apply social media data and advanced text-mining and natural language processing (NLP) techniques to identify and analyze drought impacts. ...
Acquiring a better understanding of drought impacts becomes increasingly vital under a warming climate. Traditional drought indices describe mainly biophysical variables and not impacts on social, economic, and environmental systems. We utilized natural language processing and bidirectional encoder representation from Transformers (BERT) based transfer learning to fine-tune the model on the data from the news-based Drought Impact Report (DIR) and then apply it to recognize seven types of drought impacts based on the filtered Twitter data from the United States. Our model achieved a satisfying macro-F1 score of 0.89 on the DIR test set. The model was then applied to California tweets and validated with keyword-based labels. The macro-F1 score was 0.58. However, due to the limitation of keywords, we also spot-checked tweets with controversial labels. 83.5% of BERT labels were correct compared to the keyword labels. Overall, the fine-tuned BERT-based recognizer provided proper predictions and valuable information on drought impacts. The interpretation and analysis of the model were consistent with experiential domain expertise.
... Tweets were also used to extract the damages of disasters by applying various machine-learning methods, such as support vector machine, decision tree, and logistic regression [15]. Another study compared different neural networks built on top of various word embedding models of Tweets in Hindi on detecting situational information during various disasters [16]. However, to the best of our knowledge, no study has tried to apply social media data and advanced text-mining and natural language processing (NLP) techniques to identify and analyze drought impacts. ...
Under climate change, the increasing frequency, intensity, and spatial extent of drought events lead to higher socio-economic costs. However, the relationships between the hydro-meteorological indicators and drought impacts are not identified well yet because of the complexity and data scarcity. In this paper, we proposed a framework based on the extreme gradient model (XGBoost) for Texas to predict multi-category drought impacts and connected a typical drought indicator, Standardized Precipitation Index (SPI), to the text-based impacts from the Drought Impact Reporter (DIR). The preliminary results of this study showed an outstanding performance of the well-trained models to assess drought impacts on agriculture, fire, society & public health, plants & wildlife, as well as relief, response & restrictions in Texas. It also provided a possibility to appraise drought impacts using hydro-meteorological indicators with the proposed framework in the United States, which could help drought risk management by giving additional information and improving the updating frequency of drought impacts. Our interpretation results using the Shapley additive explanation (SHAP) interpretability technique revealed that the rules guiding the predictions of XGBoost comply with domain expertise knowledge around the role that SPI indicators play around drought impacts.
... Existing content based disaster summarization approaches explore the high variance in the frequency and presence of keywords related to a disaster [61,62] to generate the summary. Additionally, several approaches initially classify each tweet either as relevant or non-relevant tweets and then, they summarize relevant tweets by semi-supervised learning [12] or supervised learning [44,58,65] on the basis of the tweet content followed by selection of representative tweets from relevant tweets by different techniques, like ILP [61], LSA [24], LUHN [42] or PageRank [52]. Additionally, recent research works utilize neural network-based techniques to utilize the tweet contents, such as Dusart et al. [15] propose utilization of BERT model [41] to identify the importance of a tweet. ...
The huge popularity of social media platforms, like Twitter, attracts a large fraction of users to share real-time information and short situational messages during disasters which belong to different topics/categories. A summary of all the messages is required by the government organizations, agencies, and volunteers for efficient and quick disaster response. Therefore, we propose an ontology based real-time tweet summarization (OntoRealSumm) for disasters which generates a summary of the tweets related to disasters with minimum human intervention. Given tweets related to a disaster, OntoRealSumm ensures fulfilment of the objectives related to disaster tweet summarization, such as identification of a category of each tweet and summarization of each category considering the category importance with respect to the disaster and ensuring information coverage of each category and diversity in summary. Comparing the performance with state-of-the-art techniques on 10 disaster datasets validates the effectiveness of OntoRealSumm.
... We consider it likely that CNNs have been adopted largely due to their capability in learning features automatically, parameter sharing and dimensionality reduction [114]. However, CNNs have performed poor for identifying word order in a sentence for text classification tasks [73]. Moreover, the computational cost (e.g., training time) for CNNs has been considerable, particularly when the training dataset is large. ...
... Although RNNs have been successful in many sequence prediction tasks, it has issues in learning longterm dependencies due to the vanishing gradient problem. This problem occurs from the gradient propagation of the recurrent network over many layers [73]. LSTM networks have been proposed to overcome these drawbacks and have shown better results for multiple text classification tasks [99]. ...
Mechanisms for sharing information in a disaster situation have drastically changed due to new technological innovations throughout the world. The use of social media applications and collaborative technologies for information sharing have become increasingly popular. With these advancements, the amount of data collected increases daily in different modalities, such as text, audio, video, and images. However, to date, practical Disaster Response (DR) activities are mostly depended on textual information, such as situation reports and email content, and the benefit of other media is often not realised. Deep Learning (DL) algorithms have recently demonstrated promising results in extracting knowledge from multiple modalities of data, but the use of DL approaches for DR tasks has thus far mostly been pursued in an academic context. This paper conducts a systematic review of 83 articles to identify the successes, current and future challenges, and opportunities in using DL for DR tasks. Our analysis is centred around the components of learning, a set of aspects that govern the application of Machine learning (ML) for a given problem domain. A flowchart and guidance for future research are developed as an outcome of the analysis to ensure the benefits of DL for DR activities are utilized.
... However, due to the development of technology, people use social media like Twitter, Facebook, etc., to exchange information that is easy to use and reached the information directly to the people. Many studies Basu et al. (2019), Rudra et al. (2018a, b), Sreenivasulu and Sridevi (2017), Madichetty and Muthukumarasamy (2020) proved that the usage of social media like Twitter to know situational awareness during the disaster. Users post millions of tweets on Twitter during the disaster. ...
In recent days, humanitarian organizations rely on social media like Twitter for situational awareness during the disaster. Millions of tweets are posted on Twitter in the form of text or images or both. Existing works showed that both image and text give complementary information during the disaster. Multi-modal informative tweet detection is helpful to both government and non-government organizations which remains a challenging task during the disaster. However, most of the existing works focused on either text or image content, but not both. In this paper, we propose a novel method based on the combination of fine-tuned BERT and DenseNet models for identifying the multi-modal informative tweets during the disaster. Fine-tuned BERT model is used to extract the linguistic, syntactic and semantic features which help deep understanding of the informative text present in the multi-modal tweet. On the other hand, the fine-tuned DenseNet model is used to extract the sophisticated features from the image. Different experiments are performed on a vast number of data-sets such as Hurricane Harvey, Hurricane Irma, Hurricane Maria, California Wildfire, Sri Lanka floods and Iraq–Iran Earthquake. Experimental results demonstrate that the proposed method outperforms the state-of-the-art method on different parameters. It is the first attempt, to the best of our knowledge, to detect multi-modal informative tweets using the combination of fine-tuned BERT and DenseNet models, where at least any text or image is informative during the disaster.
... Tweets were also used to extract the damages of disasters by applying various machine-learning methods, such as support vector machine, decision tree, and logistic regression [15]. Another study compared different neural networks built on top of various word embedding models of Tweets in Hindi on detecting situational information during various disasters [16]. However, to the best of our knowledge, no study has tried to apply social media data and advanced text-mining and natural language processing (NLP) techniques to identify and analyze drought impacts. ...
Acquiring a better understanding of drought impacts becomes increasingly vital under a warming climate. Traditional drought indices describe mainly biophysical variables and not impacts on social, economic, and environmental systems. We utilized natural language processing and bidirectional encoder representation from Transformers (BERT) based transfer learning to fine-tune the model on the data from the news-based Drought Impact Report (DIR) and then apply it to recognize seven types of drought impacts based on the filtered Twitter data from the United States. Our model achieved a satisfying macro-F1 score of 0.89 on the DIR test set. The model was then applied to California tweets and validated with keyword-based labels. The macro-F1 score was 0.58. However, due to the limitation of keywords, we also spot-checked tweets with controversial labels. 83.5% of BERT labels were correct compared to the keyword labels. Overall, the fine-tuned BERT-based recognizer provided proper predictions and valuable information on drought impacts. The interpretation and analysis of the model were consistent with experiential domain expertise.
