Automated Text Annotation for Social Media Data during Natural Disasters

Win, Si Si Mar; Aung, Than Nwe

doi:10.25046/aj030214

Cited by 12 publications

(9 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…well-known method to calculate the importance of a feature (keyword) in a given category (class) [31]. In the context of event detection, [14] used the mean PMI to select the most related features in the disaster lexicon. Moreover, KeyRate (KR) has been developed by [26] to select the most important triggers and arguments for a specific event type for event extraction tasks.…”

Section: Selection Methods Pointwise Mutual Information (Pmi) Is Amentioning

confidence: 99%

“…Nevertheless, these models are known to suffer from low generalization levels when tested on the unseen (out-of-domain) disaster event. The reason behind that is the lack of manually labeled data available to train these deep learning models, especially that they heavily rely on the quality and the quantity of the training data [11,14,22]- [24]. Our work aimed at solving this unaddressed issue.…”

Section: 11mentioning

confidence: 99%

“…The authors in [14] applied semi-supervised learning to disaster response data. They employed self-training learned on the small available data to label a new collection, Mayanmar _Earth-quake_2016, derived from Twitter with annotation accuracy of 80%.…”

Section: Introductionmentioning

confidence: 99%

“…They employed self-training learned on the small available data to label a new collection, Mayanmar _Earth-quake_2016, derived from Twitter with annotation accuracy of 80%. Similar to [14], we build a semi-supervised method, but we use distant supervision [15] via an external linguistic knowledge base, FrameNet [16], instead of self-labeling. Unlike [14], we propose a novel framework that does not duplicate the label noise exists in the current dataset and explores different ranges of unseen features by expanding the original keyword list to include new linguistic units (new keywords with similar meaning) derived from FrameNet which provides the chance to improve the generalization level of the classification model.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Automatic Labeling of Tweets for Crisis Response Using Distant Supervision

Alrashdi

O’Keefe

2020

Companion Proceedings of the Web Conference 2020

View full text Add to dashboard Cite

Current tweet classification models aimed at enhancing crisis response are based on supervised deep learning. They rely on the quality and quantity of human-labeled training data. Still, the available training data is small in size and imbalanced in coverage of crisis types, which prevents the models from generalization, and as it is manually labeled, it is also expensive to produce. To overcome these problems, distant supervision can be applied to automatically generate large-scale labeled data for tweet classification for crisis response. Experimental results on different crisis events show that our work can produce good quality labeled data from past and recent events. Substituting automatically labeled training data for part of the manually labeled training data has a minimal impact on the model performance, indicating that automatically labeled data can be used when no hand-labeled data is available.

show abstract

Section: Selection Methods Pointwise Mutual Information (Pmi) Is Amentioning

confidence: 99%

Section: 11mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automatic Labeling of Tweets for Crisis Response Using Distant Supervision

Alrashdi

O’Keefe

2020

Companion Proceedings of the Web Conference 2020

View full text Add to dashboard Cite

show abstract

“…It reaches a wider audience who would continue to spread the news until a desirable action is taken. Thus, researching this area is important to objectively reveal current events and happenings, such as breaking news, instant outbreaks, infectious disease, and terror attacks [1].…”

Section: Introductionmentioning

confidence: 99%

SMAFED: Real-Time Event Detection in Social Media Streams

Kolajo

Daramola

Adebiyi

2021

Preprint

View full text Add to dashboard Cite

Interactions via social media platforms have made it possible for anyone, irrespective of physical location, to gain access to quick information on events taking place all over the globe. However, the semantic processing of social media data is complicated due to challenges such as language complexity, unstructured data, and ambiguity. In this paper, we proposed the Social Media Analysis Framework for Event Detection (SMAFED). SMAFED aims to facilitate improved semantic analysis of noisy terms in social media streams, improved representation/embedding of social media stream content, and improved summarisation of event clusters in social media streams. For this, we employed key concepts such as integrated knowledge base, resolving ambiguity, semantic representation of social media streams, and Semantic Histogram-based Incremental Clustering based on semantic relatedness. Two evaluation experiments were conducted to validate the approach. First, we evaluated the impact of the data enrichment layer of SMAFED. We found that SMAFED outperformed other pre-processing frameworks with a lower loss function of 0.15 on the first dataset and 0.05 on the second dataset. Secondly, we determined the accuracy of SMAFED at detecting events from social media streams. The result of this second experiment showed that SMAFED outperformed existing event detection approaches with better Precision (0.922), Recall (0.793), and F-Measure (0.853) metric scores. The findings of the study present SMAFED as a more efficient approach to event detection in social media.

show abstract