A new generation of textual corpora: mining corpora from very large collections

ABSTRACT

While digital libraries based on page images and automatically generated text have made possible massive projects such as the Million Book Library, Open Content Alliance, Google, and others, humanists still depend upon textual corpora expensively produced with labor-intensive methods such as double-keyboarding and manual correction. This paper reports the results from an analysis of OCR-generated text for classical Greek source texts. Classicists have depended upon specialized manual keyboarding that costs two or more times as much as keyboarding of English both for accuracy and because classical Greek OCR produced no usable results. We found that we could produce texts by OCR that, in some cases, approached the 99.95% professional data entry accuracy rate. In most cases, OCR-generated text yielded results that, by including the variant readings that digital corpora traditionally have left out, provide better recall and, we argue, can better serve many scholarly needs than the expensive corpora upon which classicists have relied for a generation. As digital collections expand, we will be able to collate multiple editions against each other, identify quotations of primary sources, and provide a new generation of services.

REFERENCES

• 1: N. Audenaert, R. Furuta, E. Urbina, J. Deng, C. Monroy, R. Sáenz, and D. Careaga. Integrating diverse research in a digital library focused on a single author. In ECDL 05: Proceedings of the Ninth European Conference on Research and Advanced Technology for Digital Libraries, volume 3652 of Lecture Notes in Computer Science, pages 151--161. Springer, 2005.
• 2: H. S. Baird, V. Govindaraju, and D. P. Lopresti. Document analysis systems for digital libraries: challenges and opportunities. In Document Analysis Systems VI, 6th International Workshop, DAS 2004, volume 3163 of Lecture Notes in Computer Science, pages 1--16. Springer, 2004.
• 3: Regina Barzilay , Noemie Elhadad, Sentence alignment for monolingual comparable corpora, Proceedings of the 2003 conference on Empirical methods in natural language processing, p.25-32, July 11, 2003 [doi>10.3115/1119355.1119359]
• 4: A. Belaïd , I. Turcan , J. M. Pierrel , Y. Belaïd , Y. Hadjamar , H. Hadjamar, Automatic Indexing and Reformulation of Ancient Dictionaries, Proceedings of the First International Workshop on Document Image Analysis for Libraries (DIAL'04), p.342, January 23-24, 2004
• 5: J. Carlquist. Medieval manuscripts, hypertext and reading, visions of digital editions. Literary and Linguistic Computing, 19(1):105--118, 2004.
• 6: Hubert Cecotti , Abdel Belayd, Hybrid OCR combination approach complemented by a specialized ICR applied on ancient documents, Proceedings of the Eighth International Conference on Document Analysis and Recognition, p.1045-1049, August 31-September 01, 2005 [doi>10.1109/ICDAR.2005.130]
• 7: G. S. Choudhury, T. DiLauro, R. Ferguson, M. Droetthboom, and I. Fuginaga. Document recognition for a million books. D-Lib Magazine, 12(3), 2006.
• 8: G. Crane. Generating and parsing classical Greek. Literary and Linguistic Computing, 6(4):243--245, 1991.
• 9: G. Crane, D. Bamman, and A. Babeu. ePhilology: When the Books Talk to Their Readers. Blackwell Companion to Digital Literary Studies, edited by Ray Siemens and Susan Scheibman. Basil Blackwell, 2007. Forthcoming.
• 10: G. Crane, D. Bamman, L. Cerrato, A. Jones, D. Mimno, A. Packel, D. Sculley, and G. Weaver. Beyond digital incunabula: Modeling the next generation of digital libraries. In Proceedings of the 10th European Conference on Research and Advanced Technology for Digital Libraries (ECDL 2006), volume 4172 of Lecture Notes in Computer Science. Springer, 2006.
• 11: Gregory Crane , Robert F. Chavez , Anne Mahoney , Thomas L. Milbank , Jeffrey A. Rydberg-Cox , David A. Smith , Clifford E. Wulfman, Drudgery and deep thought, Communications of the ACM, v.44 n.5, p.34-40, May 2001 [doi>10.1145/374308.374333]
• 12: G. Crane and A. Jones. The Perseus American Collection 1.0. Technical report, Tufts University-Perseus Project, 2005.
• 13: Gregory Crane , Alison Jones, The challenge of virginia banks: an evaluation of named entity analysis in a 19th-century newspaper collection, Proceedings of the 6th ACM/IEEE-CS joint conference on Digital libraries, June 11-15, 2006, Chapel Hill, NC, USA [doi>10.1145/1141753.1141759]
• 14: A. Dekhtyar, I. E. Iacob, J. W. Jaromczyk, K. Kiernan, N. Moore, and D. C. Porter. Support for XML markup of image-based electronic editions. Int. J. on Digital Libraries, 6(1):55--69, 2006.
• 15: Y. Deng, S. Kumar, and W. Byrne. Segmentation and alignment of parallel text for statistical machine translation. Natural Language Engineering, 12(4):1--26, 2006.
• 16: D. F. Felluga. Addressed to the NINES: The Victorian Archive and the disappearance of the book. Victorian Studies, pages 306--319, Winter 2006.
• 17: Shaolei Feng , R. Manmatha, A hierarchical, HMM-based automatic evaluation of OCR accuracy for a digital library of books, Proceedings of the 6th ACM/IEEE-CS joint conference on Digital libraries, June 11-15, 2006, Chapel Hill, NC, USA [doi>10.1145/1141753.1141776]
• 18: B. Gatos, K. Ntzios, I. Pratikakis, S. Petridis, T. Konidaris, and S. J. Perantonis. An efficient segmentation-free approach to assist Old Greek handwritten manuscript OCR. Pattern Anal. Appl., 8(4):305--320, 2006.
• 19: H. Ghorbel, G. Coray, and A. Linden. SAM: System for multi-criteria alignment. In Proceedings of LREC 2002, 2002.
• 20: Tapas Kanungo , Philip Resnik , Song Mao , Doe-Wan Kim , Qigong Zheng, The Bible and multilingual optical character recognition, Communications of the ACM, v.48 n.6, p.124-130, June 2005 [doi>10.1145/1064830.1064837]
• 21: Yann Leydier , Frank LeBourgeois , Hubert Emptoz, Textual indexation of ancient documents, Proceedings of the 2005 ACM symposium on Document engineering, November 02-04, 2005, Bristol, United Kingdom [doi>10.1145/1096601.1096630]
• 22: Simone Marinai , Emanuele Marino , Francesca Cesarini , Giovanni Soda, A General System for the Retrieval of Document Images from Digital Libraries, Proceedings of the First International Workshop on Document Image Analysis for Libraries (DIAL'04), p.150, January 23-24, 2004
• 23: Rada Mihalcea , Michel Simard, Parallel texts, Natural Language Engineering, v.11 n.3, p.239-246, September 2005 [doi>10.1017/S1351324905003827]
• 24: David Mimno , Alison Jones , Gregory Crane, Finding a catalog: generating analytical catalog records from well-structured digital texts, Proceedings of the 5th ACM/IEEE-CS joint conference on Digital libraries, June 07-11, 2005, Denver, CO, USA [doi>10.1145/1065385.1065448]
• 25: Carlos Monroy , Rajiv Kochumman , Richard Furuta , Eduardo Urbina, Interactive Timeline Viewer (ItLv): A Tool to Visualize Variants Among Documents, Visual Interfaces to Digital Libraries [JCDL 2002 Workshop], p.39-49, January 2002
• 26: Carlos Monroy , Rajiv Kochumman , Richard Furuta , Eduardo Urbina , Eréndira Melgoza , Arpita Goenka, Visualization of Variants in Textual Collations to Analyze the Evolution of Literary Works in the Cervantes Project, Proceedings of the 6th European Conference on Research and Advanced Technology for Digital Libraries, p.638-653, September 16-18, 2002
• 27: R. Nelken and S. M. Shieber. Towards robust context-sensitive sentence alignment for monolingual corpora. In 11th Conference of the European Chapter of the Association for Computational Linguistics, 2006.
• 28: Gregory B. Newby , Charles Franks, Distributed proofreading, Proceedings of the 3rd ACM/IEEE-CS joint conference on Digital libraries, May 27-31, 2003, Houston, Texas
• 29: K. Ntzios , B. Gatos , I. Pratikakis , T. Konidaris , S. J. Perantonis, An Old Greek Handwritten OCR System, Proceedings of the Eighth International Conference on Document Analysis and Recognition, p.64-69, August 31-September 01, 2005 [doi>10.1109/ICDAR.2005.51]
• 30: C. B. Owen, J. Ford, F. Makedon, T. Steinberg, and C. Metaxaki-Kossionides. Parallel text alignment. Int. Journal of Dig. Libraries, 3(1):100--114, July 2000.
• 31: J. Ross Quinlan, C4.5: programs for machine learning, Morgan Kaufmann Publishers Inc., San Francisco, CA, 1993
• 32: S. Rawat, K. S. S. Kumar, M. Meshesha, I. D. Sikdar, A. Balasubramanian, and C. V. Jawahar. A semi-automatic adaptive OCR for digital libraries. In Document Analysis Systems VII, 7th International Workshop,, volume 3872 of Lecture Notes in Computer Science, pages 13--24, 2006.
• 33: Massimo Riva , Vika Zafrin, Extending the text: digital editions and the hypertextual paradigm, Proceedings of the sixteenth ACM conference on Hypertext and hypermedia, September 06-09, 2005, Salzburg, Austria [doi>10.1145/1083356.1083396]
• 34: P. Robinson. Where we are with electronic scholarly editions, and where we want to be. Jahrbuch fr Computerphilogie - online, 5:123--43, 2004.
• 35: Desmond Schmidt , Theodor Wyeld, A novel user interface for online literary documents, Proceedings of the 19th conference of the computer-human interaction special interest group (CHISIG) of Australia on Computer-human interaction: citizens online: considerations for today and the future, November 21-25, 2005, Canberra, Australia
• 36: S. Schreibman, A. Kumar, and J. McDonald. The versioning machine. Literary and Linguistic Computing, 18(1):101--7, 2003.
• 37: D. A. Smith. Textual variation and version control in the TEI. Computers and the Humanities, 33(1--2):103--112, April 1999.
• 38: David A. Smith , Gregory Crane, Disambiguating Geographic Names in a Historical Digital Library, Proceedings of the 5th European Conference on Research and Advanced Technology for Digital Libraries, p.127-136, September 04-09, 2001
• 39: M. Spencer, B. Bordalejo, L. Wang, A. Barbrook, L. Mooney, P. Robinson, T. Warnow, and C. Howe. Analyzing the order of items in manuscripts of the Canterbury Tales. Computers and the Humanities, 37(1):97--109, February 2003.
• 40: M. Spencer and C. Howe. Collating texts using progressive multiple alignment. Computers and the Humanities, 38(3):253--70, August 2004.
• 41: S. F. Thomas. Finalizing the multiple-text electronic King Lear for use in the classroom. Proceedings of ACH/ALLC 2005, Victoria, 15-18 Jun 2005, 2005.
• 42: J. Veronis. Parallel text processing: Alignment and use of translation corpora. Kluwer Academic Publishers, 2000.
• 43: Arkady B. Zaslavsky , Alejandro Bia , Krisztián Monostori, Using Copy-Detection and Text Comparison Algorithms for Cross-Referencing Multiple Editions of Literary Works, Proceedings of the 5th European Conference on Research and Advanced Technology for Digital Libraries, p.103-114, September 04-09, 2001

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.7 Digital Libraries

General Terms:
Documentation, Measurement

Keywords:
OCR evaluation, ancient greek, text alignment