Stacked Sentence-Document Classifier Approach for Improving Native Language Identification.

Andrea Cimino Felice Dell'Orletta

Published in: BEA@EMNLP (2017)

Keyphrases

language identification
document images
word level
document analysis
natural language
training data
indian languages
information retrieval systems
document collections
feature selection
information retrieval
retrieval systems
text documents
optical character recognition
hough transform
multi modal
feature set