Conference paper Open Access
Leveraging Transformer Self Attention Encoder for Crisis Event Detection in Short Texts
Pantelis Kyriakidis;
Despoina Chatzakou;
Theodora Tsikrika;
Stefanos Vrochidis;
Ioannis Kompatsiaris
Analyzing content generated on social media has proven to be a powerful tool for early detection of crisis-related events. Such an analysis may allow for timely action, mitigating or even preventing altogether the effects of a crisis. However, the high noise levels in short texts present in microblogging platforms, combined with the limited publicly available datasets have rendered the task difficult. Here, we propose deep learning models based on a transformer self-attention encoder, which is capable of detecting event-related parts in a text, while also minimizing potential noise levels. Our models efficacy is shown by experimenting with CrisisLexT26, achieving up to 81.6% f1-score and 92.7% AUC.
| Name | Size | |
|---|---|---|
|
Leveraging Transformer Self Attention Encoder.pdf
md5:bdcb24641493d1f4408f17285c5189bd |
372.4 kB | Download |
| All versions | This version | |
|---|---|---|
| Views | 168 | 168 |
| Downloads | 151 | 151 |
| Data volume | 56.2 MB | 56.2 MB |
| Unique views | 145 | 145 |
| Unique downloads | 134 | 134 |
- Publication date:
- February 10, 2022
- DOI:
-
- Keyword(s):
- Self attention Multihead attention Crisis event detection
- Grants:
-
European Commission:
-
INFINITY - IMMERSE. INTERACT. INVESTIGATE (883293)
- AIDA - Artificial Intelligence and advanced Data Analytics for Law Enforcement Agencies (883596)
-
INFINITY - IMMERSE. INTERACT. INVESTIGATE (883293)
- Meeting:
- 44th European Conference on Information Retrieval (ECIR'22), Stavanger, Norway, 10-14 April 2022
- Communities:
- License (for files):
- Creative Commons Attribution 2.0 Generic