Login

Login
Welcome:
Guest

Search for:


Browse:

Bannner: Aslib individual membership.
 
Journal search
Journal cover: Aslib Proceedings

Aslib Proceedings

ISSN: 0001-253X

Online from: 1949

Subject Area: Library and Information Studies

Content: Latest Issue | icon: RSS Latest Issue RSS | Previous Issues

Options: To add Favourites and Table of Contents Alerts please take a Emerald profile

Previous article.Icon: Print.Table of Contents.Next article.Icon: .

Interpreting the image: using advanced computational techniques to read the Vindolanda texts


Document Information:
Title:Interpreting the image: using advanced computational techniques to read the Vindolanda texts
Author(s):Melissa Terras, (School of Library, Archive and Information Studies, University College London, London, UK)
Citation:Melissa Terras, (2006) "Interpreting the image: using advanced computational techniques to read the Vindolanda texts", Aslib Proceedings, Vol. 58 Iss: 1/2, pp.102 - 117
Keywords:Artificial intelligence, Historical research, Image processing, Knowledge processes
Article type:Research paper
DOI:10.1108/00012530610648707 (Permanent URL)
Publisher:Emerald Group Publishing Limited
Abstract:

Purpose – This paper aims to provide an overview of the development of a computer system designed to aid historians in the reading of the stylus tablets from the Roman fort of Vindolanda. It proposes outlining the different stages in developing the system, and giving the preliminary results.

Design/methodology/approach – The paper provides a literature review regarding Vindolanda, stylus tablets, and the process of reading an ancient document. Knowledge elicitation techniques are used to model explicitly expert processes used to read an ancient document. A corpus of character forms and lexicostatistics is gathered. An advanced cognitive imaging system utilising artificial intelligence techniques is implemented to produce plausible interpretations of the document.

Findings – This paper describes the developmental stages undertaken to construct a system that can read in images of an ancient document and produce plausible interpretations of the document, to aid the historians in the lengthy process of reading an ancient text. In carrying out the development, an explicit representation of how experts approach and reason about damaged and deteriorated texts was formulated, and a large corpus of letter forms and linguistic data were captured. Preliminary results from the resulting computer system are presented which demonstrate the usefulness of the technique, although more work is needed to develop this into a stand-alone computer system.

Research limitations/implications – The study is focused on the Roman stylus tablets from Vindolanda, near Hadrian's Wall, although the technique could be extrapolated to cover other types of ancient documents from any period.

Practical implications – It is demonstrated that using techniques from artificial intelligence and cognitive psychology can result in further explicit understanding of humanities expert processes, which allow computational systems to be constructed. The resulting computational system is a tool for the humanities expert, which carries out a task in a similar manner, allowing for faster reasoning time and quicker hypotheses development.

Originality/value – The paper presents the first known system to intake an image of an ancient text and output a plausible interpretation of the text in a reasonable time frame, assisting the papyrologist in resolving ambiguities in the damaged and abraded text.



Fulltext Options:

Login

Login

Existing customers: login
to access this document

Login


- Forgot password?

- Athens/Institutional login

Purchase

Purchase

Downloadable; Printable; Owned
HTML, PDF (501kb)Purchase

To purchase this item please login or register.

Login


- Forgot password?

Recommend to your librarian

Complete and print this form to request this document from your librarian


Marked list

Bookmark & share

Reprints & permissions

© Emerald Group Publishing Limited  |  Copyright information  |  Site policies  |  Cookie information
.