AraDIC: Arabic Document Classification using Image-Based Character Embeddings and Class-Balanced Loss.

Mahmoud Daif Shunsuke Kitada Hitoshi Iyatomi

Published in: CoRR (2020)

Keyphrases

document classification
text classification
text categorization
text mining
text documents
web documents
classification algorithm
topic extraction
data sets
machine learning
object recognition
knn
k nearest neighbor
n gram
optical character recognition