A Flexible Model for Enterprise Document Capturing Automation

Juris Rāts, Inguna Pede, Tatjana Rubina, Gatis Vītols

2020

Abstract

The aim of the research is to create and evaluate a flexible model for document capturing that would employ machine learning to classify documents feeding them with values for one or more metadata items. Documents and classification metadata fields typical for Enterprise Content Management (ECM) systems are used in the research. The model comprises selection of classification methods, configuration of the methods hyperparameters and configuration of a number of other learning related parameters. The model provides user with visual means to analyse the classification outcomes and those to tune the further steps of the learning. A couple of challenges are addressed along the way – as informal and eventually changing criteria for document classification, and imbalanced data sets.

Paper Citation

in Harvard Style

Rāts J., Pede I., Rubina T. and Vītols G. (2020). A Flexible Model for Enterprise Document Capturing Automation.In Proceedings of the 22nd International Conference on Enterprise Information Systems - Volume 1: ICEIS, ISBN 978-989-758-423-7, pages 297-304. DOI: 10.5220/0009034802970304

in Bibtex Style

@conference{iceis20,
author={Juris Rāts and Inguna Pede and Tatjana Rubina and Gatis Vītols},
title={A Flexible Model for Enterprise Document Capturing Automation},
booktitle={Proceedings of the 22nd International Conference on Enterprise Information Systems - Volume 1: ICEIS,},
year={2020},
pages={297-304},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0009034802970304},
isbn={978-989-758-423-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 22nd International Conference on Enterprise Information Systems - Volume 1: ICEIS,
TI - A Flexible Model for Enterprise Document Capturing Automation
SN - 978-989-758-423-7
AU - Rāts J.
AU - Pede I.
AU - Rubina T.
AU - Vītols G.
PY - 2020
SP - 297
EP - 304
DO - 10.5220/0009034802970304