Exploring multi-tasking learning in document attribute classification

Mondal, Tanmoy; Das, Abhijit; Ming, Zuheng

doi:10.1016/j.patrec.2022.02.015

Cited by 6 publications

(3 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This section exists independently in this review because it is now the subject of study within the field of FDE as examiners are called upon more and more to examine digital documents. In the period under review nine articles [ [647] , [648] , [649] , [650] , [651] , [652] , [653] , [654] , [655] ], 13 conference presentations [ [656] , [657] , [658] , [659] , [660] , [661] , [662] , [663] , [664] , [665] , [666] , [667] , [668] ], and five workshops [ [669] , [670] , [671] , [672] , [673] ] were dedicated to this subject.…”

Section: Forensic Document Examinationmentioning

confidence: 99%

“…In this study [ 654 ], Mondal et al explored a Multi-Tasking learning (MTL) based network to perform document attribute classification They used a MTL based network for the classification of a full document image, based on segmented word images and patches. such as the font type, font size, font emphasis and scanning resolution classification of a document image.…”

Section: Forensic Document Examinationmentioning

confidence: 99%

See 1 more Smart Citation

Interpol questioned documents review 2019–2022

Deviterne-Lapeyre

Ibrahim²

2023

Forensic Science International: Synergy

View full text Add to dashboard Cite

Section: Forensic Document Examinationmentioning

confidence: 99%

Section: Forensic Document Examinationmentioning

confidence: 99%

Interpol questioned documents review 2019–2022

Deviterne-Lapeyre

Ibrahim²

2023

Forensic Science International: Synergy

View full text Add to dashboard Cite

“…In [24], the authors focused on the classification of document attributes in a large collection of documents (scientific articles of single or double column, programming code, novels, legal texts, etc.) using only their scanned images and a deep neural network architecture.…”

Section: Fusion Of Text and Imagesmentioning

confidence: 99%

Classification of Scientific Documents in the Kazakh Language Using Deep Neural Networks and a Fusion of Images and Text

Bogdanchikov

Ayazbayev

Varlamis

2022

BDCC

View full text Add to dashboard Cite

The rapid development of natural language processing and deep learning techniques has boosted the performance of related algorithms in several linguistic and text mining tasks. Consequently, applications such as opinion mining, fake news detection or document classification that assign documents to predefined categories have significantly benefited from pre-trained language models, word or sentence embeddings, linguistic corpora, knowledge graphs and other resources that are in abundance for the more popular languages (e.g., English, Chinese, etc.). Less represented languages, such as the Kazakh language, balkan languages, etc., still lack the necessary linguistic resources and thus the performance of the respective methods is still low. In this work, we develop a model that classifies scientific papers written in the Kazakh language using both text and image information and demonstrate that this fusion of information can be beneficial for cases of languages that have limited resources for machine learning models’ training. With this fusion, we improve the classification accuracy by 4.4499% compared to the models that use only text or only image information. The successful use of the proposed method in scientific documents’ classification paves the way for more complex classification models and more application in other domains such as news classification, sentiment analysis, etc., in the Kazakh language.

show abstract