Semantic Wide and Deep Learning for Detecting Crisis-Information Categories on Social Media

@inproceedings{Burel2017SemanticWA,
  title={Semantic Wide and Deep Learning for Detecting Crisis-Information Categories on Social Media},
  author={Gr{\'e}goire Burel and Hassan Saif and Harith Alani},
  booktitle={SEMWEB},
  year={2017}
}
When crises hit, many flog to social media to share or consume information related to the event. [] Key Method Unlike previous models, which mainly rely on the lexical representations of words in the text, the proposed model integrates an additional layer of semantics that represents the named entities in the text, into a wide and deep CNN network. Results show that the Sem-CNN model consistently outperforms the baselines which consist of statistical and non-semantic deep learning models.
View via Publisher
oro.open.ac.uk
51 Citations
Highly Influential Citations
3
Background Citations
24
Methods Citations
15
Results Citations
2

51 Citations

Citation Type

Citation Type

Crisis Event Extraction Service (CREES) - Automatic Detection and Classification of Crisis-related Content on Social Media
TLDR
The Crisis Event Extraction Service (CREES), an open-source web API that automatically classifies posts during crisis situations, is introduced and results show that the CNN-based API results can be relied upon when dealing with specific crises with the benefits associated with the usage word embeddings.
I-AID: Identifying Actionable Information From Disaster-Related Tweets
TLDR
This paper proposes a system, dubbed I-AID, to automatically filter tweets with critical or actionable information from the enormous volume of social media data, and combines state-of-the-art approaches to process and represents textual data in order to capture its underlying semantics.
Classifying Crises-Information Relevancy with Semantics
TLDR
The impact of semantics in classifying Twitter posts across same, and different, types of crises is explored, showing that adding semantic features has no noticeable benefit over statistical features when classifying same-type crises, whereas it enhances the classifier performance by up to 7.2% when classifies information about a new type of crisis.
Classifying Crises-Information Relevancy with Semantics Conference or Workshop Item
TLDR
The impact of semantics in classifying Twitter posts across same, and different, types of crises is explored, showing that adding semantic features has no noticeable benefit over statistical features when classifying same-type crises, whereas it enhances the classifier performance by up to 7.2% when classifies information about a new type of crisis.
Multi-source Multimodal Data and Deep Learning for Disaster Response: A Systematic Review
TLDR
A systematic review of 83 articles to identify the successes, current and future challenges, and opportunities in using DL for DR tasks, centred around the components of learning, a set of aspects that govern the application of Machine learning for a given problem domain.
Cross-Lingual Classification of Crisis Data
TLDR
This paper test statistical and semantic classification approaches on cross-lingual datasets from 30 crisis events, consisting of posts written mainly in English, Spanish, and Italian, and shows that the addition of semantic features extracted from external knowledge bases improve accuracy over a purely statistical model.
A Novel Framework for Detecting Important Subevents from Crisis Events via Dynamic Semantic Graphs
TLDR
This work proposes a novel framework to extract critical sub-events from a large-scale crisis event by combining important information across relevant tweets and shows that this approach significantly outperforms event detection baselines.
DICE @ TREC-IS 2018: Combining Knowledge Graphs and Deep Learning to Identify Crisis-Relevant Tweets
TLDR
The TREC-IS results indicate that a model based on combining knowledge graphs, word embeddings and textual features outperformes classical machine learning models.
Identifying and Processing Crisis Information from Social Media
TLDR
The results obtained through various analyses in this thesis demonstrate the value of semantic enrichment of text through knowledge graphs in improving the adaptability of crisis relevancy classifiers across crisis types and languages, in comparison to statistical features as often used in much of the related work.
Few-shot tweet detection in emerging disaster events
TLDR
This work compares how few-shot approaches (matching networks and prototypical networks) perform for this task and demonstrates how a modified one-class version of prototypical models can be used for this application.
...
...

References

SHOWING 1-10 OF 30 REFERENCES
On Semantics and Deep Learning for Event Detection in Crisis Situations
TLDR
Dual-CNN is introduced, a semantically-enhanced deep learning model to target the problem of event detection in crisis situations from social media data that is competitive with more traditional Machine Learning models, such as SVM.
Event Extraction via Dynamic Multi-Pooling Convolutional Neural Networks
TLDR
A word-representation model to capture meaningful semantic regularities for words and a framework based on a convolutional neural network to capture sentence-level clues are introduced.
A Semantic Graph-Based Approach for Radicalisation Detection on Social Media
TLDR
The approach extracts and makes use of the underlying semantics of words exhibited by Twitter users to identify their pro/anti-ISIS stances and results show that classifiers trained from semantic features outperform those trained from lexical, sentiment, topic and network features.
Deep Convolutional Neural Networks for Sentiment Analysis of Short Texts
TLDR
A new deep convolutional neural network is proposed that exploits from characterto sentence-level information to perform sentiment analysis of short texts and achieves state-of-the-art results for single sentence sentiment prediction.
Document Modeling with Gated Recurrent Neural Network for Sentiment Classification
TLDR
A neural network model is introduced to learn vector-based document representation in a unified, bottom-up fashion and dramatically outperforms standard recurrent neural network in document modeling for sentiment classification.
A language-independent neural network for event detection
TLDR
A language-independent neural network is developed to capture both sequence and chunk information from specific contexts and use them to train an event detector for multiple languages without any manually encoded features.
A Survey of Techniques for Event Detection in Twitter
TLDR
A survey of techniques for event detection from Twitter streams aimed at finding real‐world occurrences that unfold over space and time and highlights the need for public benchmarks to evaluate the performance of different detection approaches and various features.
Social Sensing for Urban Crisis Management: The Case of Singapore Haze
TLDR
This paper adopts a comprehensive social event analysis framework covering content, emotion, activity, and network and proposes a set of measures for each dimension accordingly to help public and private sectors to prepare themselves for future haze related events.
Benchmarking the Extraction and Disambiguation of Named Entities on the Semantic Web
TLDR
This work presents an approach that combines the state-of-the art from named entity recognition in the natural language processing domain and named entity linking from the semantic web community and relies on the numerous web extractors supported by the NERD framework to optimize recognition and linking of named entities.
Convolutional Neural Networks for Sentence Classification
TLDR
The CNN models discussed herein improve upon the state of the art on 4 out of 7 tasks, which include sentiment analysis and question classification, and are proposed to allow for the use of both task-specific and static vectors.
...
...