ConTextING: Granting Document-Wise Contextual Embeddings to Graph Neural Networks for Inductive Text Classification.
Yen-Hao HuangYi-Hsin ChenYi-Shin ChenPublished in: COLING (2022)
Keyphrases
- text classification
- neural network
- document classification
- text documents
- text classifiers
- term frequency
- training documents
- machine learning
- automatic text classification
- bag of words
- vector space
- topic discovery
- text categorization
- pattern recognition
- text mining
- labeled data
- n gram
- graph theory
- feature selection
- contextual information
- directed graph
- web documents
- random walk
- document representation
- text data
- inductive learning
- classify documents
- information retrieval
- document clustering
- artificial neural networks
- pairwise
- naive bayes
- data cleaning
- knn
- structured data
- knowledge representation
- manifold learning
- context sensitive
- euclidean space
- data points
- document collections
- keywords
- semi supervised learning
- knowledge discovery