10th Annual International Conference on Industrial Engineering and Operations Management

Evaluation for Tools and Techniques for Text-Extraction

0 Paper Citations
1 Views
1 Downloads
Track: Design and Analysis
Abstract

Sustainability and Corporate Social Responsibility (CSR) has become important not only to corporations but to society
and government as well. As a result, there has been a dramatic increase on how these corporations are informing and
disclosing their social and environmental activities in which they are involved through their company annual reports.
These annual reports are not only useful for companies but also to government, NGO’s, society and many other
stakeholders. In the past many of the annual reports were studied manually. But now with internet, there is a great
opportunity to develop automation through software tools and techniques. Data Mining techniques and related
techniques and softwares can help not only researchers but also corporations to extract text from multiple data sources
like word processing documents as well as image documents like an image file or a PDF file. In this paper, the
researchers first and designed the manual process and also performed the manual process of text extraction from
annual reports. Having understood and performed the manual process the researchers analyzed the various tools and techniques available on net. They explored and analyze all the tools, scripts and software available on the net. The
researchers analyzed through various parameters. The researchers developed a benchmarking model across all the
tools and techniques after analyzing each of these software critically. These softwares were specifically analyzed for
performance and feasibility and various parameters were evaluated within performance and feasibility.

Published in: 10th Annual International Conference on Industrial Engineering and Operations Management, Dubai, United Arab Emirates

Publisher: IEOM Society International
Date of Conference: March 10-12, 2020

ISBN: 978-1-5323-5952-1
ISSN/E-ISSN: 2169-8767