COURAGE at CheckThat!-2022: Harmful Tweet Detection using Graph Neural Networks and ELECTRA.
Francesco LomonacoGregor DonabauerMarco SiinoPublished in: CLEF (Working Notes) (2022)
Keyphrases
- neural network
- pattern recognition
- false positives
- random walk
- detection accuracy
- graph model
- fuzzy logic
- multi layer
- automatic detection
- social media
- object detection
- false alarms
- graph matching
- connected components
- graph theory
- detection rate
- sentiment analysis
- named entities
- back propagation
- detection algorithm
- machine learning
- directed graph
- neural nets
- bipartite graph
- detection method
- object recognition