The impact of deep learning on document classification using semantically rich representations

Highlights：

• Provides a novel document representation model enriched with semantical information.

• Make use of the background knowledge extracted from an ontology for incorporating semantics.

• An in-depth analysis and accuracy measurement of document classification using multiple deep learning architecture configurations.

• A comprehensive comparative evaluation of conventional machine learning and deep learning techniques on document classification.

• Proposed deep learning techniques outperform the conventional ones, achieving better accuracy in every point of testing.