Similarity-Based Synthetic Document Representations for Meta-Feature Generation in Text Classification.
Sérgio D. CanutoThiago SallesThierson Couto RosaMarcos André GonçalvesPublished in: SIGIR (2019)
Keyphrases
- feature generation
- text classification
- document representation
- text categorization
- bag of words
- text documents
- text data
- semantic features
- document classification
- feature selection
- document clustering
- text mining
- machine learning
- semantic information
- inductive learning
- n gram
- labeled data
- document collections
- knn
- k nearest neighbor
- web documents
- data fusion
- information extraction
- unlabeled data
- unsupervised learning
- image classification
- language model
- action recognition
- inductive logic programming
- word sense disambiguation
- vector space model
- nearest neighbor
- artificial intelligence