ViBERTgrid: A Jointly Trained Multi-modal 2D Document Representation for Key Information Extraction from Documents.
Weihong LinQifang GaoLei SunZhuoyao ZhongKai HuQin RenQiang HuoPublished in: ICDAR (1) (2021)
Keyphrases
- multi modal
- document representation
- information extraction
- text documents
- web documents
- document collections
- document clustering
- bag of words
- document content
- vector space model
- text mining
- document categorization
- index terms
- language model
- text data
- vector space
- information retrieval
- data fusion
- text classification
- high dimensional
- structured data
- semantic information
- precision and recall
- question answering
- named entities
- background knowledge
- natural language processing
- text categorization
- training set
- machine learning
- information retrieval systems
- data mining
- wordnet
- semantic relations
- image classification
- keywords