Muutke küpsiste eelistusi

E-raamat: Image to Interpretation: An Intelligent System to Aid Historians in Reading the Vindolanda Texts [Oxford Scholarship Online e-raamatud]

(Lecturer in Electronic Communication, School of Library, Archive and Information Studies, University College London)
Teised raamatud teemal:
  • Oxford Scholarship Online e-raamatud
  • Raamatu hind pole hetkel teada
Image to Interpretation: An Intelligent System to Aid Historians in Reading the Vindolanda TextsSuurem pilt
Teised raamatud teemal:
Wiley OnlineRohkem infot Oxford Scholarship Online e-raamatute kohta
Raamatu kodulehekülg: http://www.oxfordscholarship.com/view/10.1093/acprof:oso/9780199204557.001.0001/acprof-9780199204557
The ink and stylus tablets discovered at the Roman fort of Vindolanda are a unique resource for scholars of ancient history. However, the stylus tablets in particular are extremely difficult to read. This book details the development of what appears to be the first system constructed to aid experts in the process of reading an ancient document, exploring the extent to which techniques from Artificial Intelligence can be used to develop a system that could aid historians in reading the stylus texts. Image to Interpretation includes a model of how experts read ancient texts, a corpora of letter forms from the Vindolanda text corpus, and a detailed description of the architecture of the system. It will be of interest to papyrologists, researchers in Roman history and palaeography, computer and engineering scientists working in the field of Artificial Intelligence and image processing, and those interested in the use of computing in the humanities.
List of Tables xii
1. Introduction 1
1.1 Vindolanda
2
1.2 The Vindolanda Texts
5
1.3 Image-Processing Techniques
9
1.4 An Interface for the System
12
1.5 Papyrology and Computing
12
1.6 Research Approach
17
1.7
Chapter Overview
22
I. Knowledge Elicitation and the Papyrologist 25
2. How do Papyrologists Read Ancient Texts?
27
2.1 Papyrology Discussed
28
2.2 Knowledge Elicitation: A Brief Guide
38
2.3 Knowledge Elicitation and Vindolanda
41
2.4 Associated Techniques: Content and Textual Analysis
47
2.5 Results
51
2.6 General Observations
72
2.7 Preliminary Conclusions
75
2.8 Models of Reading and Papyrology
77
2.9 Conclusion
82
3. The Palaeography of Vindolanda
84
3.1 Palaeography
85
3.2 Knowledge Elicitation and Palaeography
91
3.3 Information Used When Discussing Letter Forms
93
3.4 Derived Encoding Scheme
98
3.5 Building the Data Set
101
3.6 The Use of the GRAVA Annotator
105
3.7 Annotating the Characters
107
3.8 Results
110
3.9 Gathering Corpus Textual Data
111
3.10 How Representative is the Corpus?
113
3.11 Letter Forms
115
3.12 Conclusion
118
II. Image to Interpretation 121
4. Using Artificial Intelligence to Read the Vindolanda Texts
co-authored with Dr Paul Robertson
123
4.1 Introduction to the GRAVA System
124
4.2 The Original System Demonstrated
126
4.3 Preliminary Experiments
129
4.4 The Construction of Character Models
132
4.5 The New System Explained
133
4.6 An Agent In Action
135
4.7 Conclusion
137
5. Results
co-authored with Dr Paul Robertson
138
5.1 Using Ink Data for Ink Tablets
139
5.2 Using Ink Models for an Unknown Phrase
143
5.3 Using Ink Models for Stylus Tablets
145
5.4 Using Stylus Models for Stylus Tablets
147
5.5 Analysing Automated Data
148
5.6 Future Work
149
5.7 Conclusion
151
6. Conclusion
153
6.1 Knowledge Elicitation
154
6.2 Expanding the Existing System
158
6.3 Application Development
163
6.4 Evaluation
165
6.5 The Wider Scope
168
6.6 In Conclusion
169
Appendix A. Using a Stochastic MDL Architecture to Read the Vindolanda Texts 171
co-authored with Dr Paul Robertson
Appendix B. Annotation 200
Appendix C. Vindolanda Letter Forms Corpus 211
References 231
Index 249


Melissa Terras is Lecturer in Electronic Communication, School of Library, Archive and Information Studies, University College London.
Püsilink: https://www.kriso.ee/db/9780199204557_pe.html
Märksõnad: