Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

Extraction and Classification of Financial Documents: State of the Art, Methodology and Perspectives

Domain:

natural language processingsocioeconomic

Record type:

paper
Creator:
FouSadJul
Publisher:
IST
Host:
Supreme Audit Institutions (SAIs) are crucial for transparency in public financial management but face major challenges when manually processing large volumes of multilingual, semi-structured, and heterogeneous financial documents. This paper reviews recent advances in multimodal Transformers and OCR-free models for document artificial intelligence and proposes a modular and reproducible pipeline for financial document analysis in African SAI contexts. The pipeline integrates layout-aware models (LayoutLMv3, DocFormer) and OCR-free architectures (Donut) and is validated on public benchmarks and a proprietary Francophone dataset (NOVAFRIQ). Experimental results show up to 93.7% F1-score for information extraction and 95.2% accuracy for document classification. Pilot simulations indicate potential processing time reductions of 60–75% and error decreases of approximately 40%. The study demonstrates a promising approach for building robust, explainable, and multilingual AI-assisted auditing tools.

Visit

doi.org

Tasks

computer visioninformation extractionoptical character recognitiontext classification

Similar

Classification of Arabic script using multiple sources of information: State of the art and perspectivesIntroducing a state-of-the-art phylogenetic classification of the Kikongo Language ClusterDeep Learning methods for biotic and abiotic stresses detection in fruits and vegetables: state of the art and perspectivesQuantitative Information Extraction from Humanitarian DocumentsCuckoo Search: State-of-the-Art and OpportunitiesSémiotique à l’université algérienne Semiotics in Algerian Academia État des lieux et perspectives State of the Art and Prospects

Classification of Arabic script using multiple sources of information: State of the art and perspectives

Introducing a state-of-the-art phylogenetic classification of the Kikongo Language Cluster

It is shown how past lexicostatistic efforts eventually led to lexically-driven phylogenetic classif

Deep Learning methods for biotic and abiotic stresses detection in fruits and vegetables: state of the art and perspectives

Deep Learning methods for biotic and abiotic stresses detection in fruits and vegetables: state of the art and perspectives

Poster presented at the Deep Learning Indaba 2022 by Ariane HOUETOHOSSOU

Quantitative Information Extraction from Humanitarian Documents

Humanitarian action is accompanied by a mass of reports, summaries, news, and other documents. To gu

Cuckoo Search: State-of-the-Art and Opportunities

Since the development of cuckoo search (CS) by Yang and Deb in 2009, CS has been applied in a divers

Sémiotique à l’université algérienne Semiotics in Algerian Academia État des lieux et perspectives State of the Art and Prospects

A la différence de son histoire en Europe, la sémiotique demeure une discipline relativement nouvell