Machine Learning based Classification of Online News Data for Disaster Management

Gopal, Lakshmi; Prabha, R.; Pullarkatt, Divya; Ramesh, Maneesha Vinodini

doi:10.1109/ghtc46280.2020.9342921

Cited by 17 publications

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The experiment used the Trafficking-10k dataset to demonstrate the efficiency of the proposed approach for data preparation in a model creation. The results were then compared to those obtained from models used in other works [25]- [27]. The Trafficking-10k dataset, which was labeled as the seven levels as shown in Table 1, was relabeled into two classes: Class 0 representing levels 0-2 (not related trafficking) and Class 1 representing levels 4-6 (related trafficking).…”

Section: Methodsmentioning

confidence: 99%

Improving misspelled word solving for human trafficking detection in online advertising data

Wiriyakun,

Kurutach

2023

IJECE

View full text Add to dashboard Cite

<span lang="EN-US">Social media is used by pimps to advertise their businesses for adult services due to easy accessibility. This requires the potentially computational model for law enforcement authorities to facilitate a detection of human trafficking activities. The machine learning (ML) models used to detect these activities mostly rely on text classification and often omit the correction of misspelled words, resulting in the risk of predictions error. Therefore, an improvement data processing approach is one of strategies to enhance an efficiency of human trafficking detection. This paper presents a novel approach to solving spelling mistakes. The approach is designed to select misspelled words, the replace them with the popular words having the same meaning based on an estimation of the probability of words and context used in human trafficking advertisements. The applicability of the proposed approach was demonstrated with the labeled human trafficking dataset using three classification models: k-nearest neighbor (KNN), naive Bayes (NB), and multilayer perceptron (MLP). The achievement of higher accuracy of the model predictions using the proposed method evidences an improved alert on human trafficking outperforming than the others. The proposed approach shows the potential applicability to other datasets and domains from the online advertisements.</span>

show abstract

Section: Methodsmentioning

confidence: 99%

Improving misspelled word solving for human trafficking detection in online advertising data

Wiriyakun,

Kurutach

2023

IJECE

View full text Add to dashboard Cite

show abstract

“…In [ 12 ] the authors utilized online news to monitor disasters. Machine learning-based techniques have been used in their work to detect useful data using irrelevant data filtration.…”

Section: Related Workmentioning

confidence: 99%

“…To train a robust deep learning model, the training process should make use of large datasets. To utilize the new advances proposed by deep learning in disaster detection and rescue efforts, several disaster datasets have been proposed in the English language [12][13][14][15][16][17][18]. However, there is a relative shortage of similar datasets for the Arabic language.…”

Section: Arbertmentioning

confidence: 99%

Enhanced Arabic disaster data classification using domain adaptation

Moussa,

Abdou,

Elsayed

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

Natural disasters, like pandemics and earthquakes, are some of the main causes of distress and casualties. Governmental crisis management processes are crucial when dealing with these types of problems. Social media platforms are among the main sources of information regarding current events and public opinion. So, they have been used extensively to aid disaster detection and prevention efforts. Therefore, there is always a need for better automatic systems that can detect and classify disaster data of social media. In this work, we propose enhanced Arabic disaster data classification models. The suggested models utilize domain adaptation to provide state-of-the-art accuracy. We used a standard dataset of Arabic disaster data collected from Twitter for testing the proposed models. Experimental results show that the provided models significantly outperform the previous state-of-the-art results.

show abstract

“…", we analyse the articles listed in the Social Media Literature Database under the theme of "Disaster Management". By studying various articles (Palen et al, 2010;Sakaki et al, 2010;Zhou et al, 2013;Jongman et al, 2015;Musaev et al, 2018;Phengsuwan et al, 2019;Guntha et al, 2020b;Gopal et al, 2020;Guntha et al, 2020a;Gopal et al, 2022;Aswathy et al, 2022) and based on our experience, we have defined nine generic actionable information categories which will be assigned to each article under the "Disaster management" theme listed in the literature database. 1 displays the various categories with their respective descriptions, detailing the methods and applications considered within each AI category in this study.…”

Section: Actionable Information (Ai) Analysismentioning

confidence: 99%

Review Article: Leveraging Social Media for Managing Natural Hazard Disasters: A Critical Review of Data Collection Strategies and Actionable Insights

Gopal,

Prabha,

Thirugnanam

et al. 2024

Preprint

View full text Add to dashboard Cite

Abstract. This paper critically reviews 250 articles from 2010 to September 2023, analyzing how social media data is utilized in disaster management, addressing challenges in relevance filtering and noise reduction to extract actionable disaster information and enhance decision-making efficiency. The results of our critical analysis are given in a Social Media Literature Database where we categorize each article's information into 7 main categories and 27 subcategories, covering article details, case study regions, disaster events, social media data specifics, data collection and analysis methods, and evaluation methodologies. To assess the effectiveness of social media in providing actionable disaster information, we further classify the articles into 9 categories, covering public discourse analysis, temporal and spatial insights, relevance filtering methods, community/stakeholder collaborations, disaster trends, and resource identification. We also illuminate historical disaster events within the review period and discuss the results through graphical visualizations. Our findings show that natural language processing methods, particularly content analysis, were commonly utilized in the literature, and contribute significantly to basic data filtering by removing noise. Commonly used advanced robust analysis machine learning methods included Support Vector Machines, Naive Bayes, and Neural Networks. We found that proficiency in temporal and spatial analysis of social media data is widespread among the studies, with varying success in implementing effective relevance filtering. Our actionable information categorization revealed a need for further exploration into community interactions and resource identification using social media data during and after disasters. Based on the literature study and our own experience on the subject, we propose six best practices for social media usage in disaster situations for the community and five best practices for researchers to enhance disaster management strategies.

show abstract

Machine Learning based Classification of Online News Data for Disaster Management

Cited by 17 publications

References 7 publications

Improving misspelled word solving for human trafficking detection in online advertising data

Improving misspelled word solving for human trafficking detection in online advertising data

Enhanced Arabic disaster data classification using domain adaptation

Review Article: Leveraging Social Media for Managing Natural Hazard Disasters: A Critical Review of Data Collection Strategies and Actionable Insights

Contact Info

Product

Resources

About