Abstract
The retrieval of OCR degraded text using n-gram formulations within a probabilistic retrieval system is examined in this paper. Direct retrieval of documents using n-gram databases of 2 and 3-grams or 2, 3, 4 and 5-grams resulted in improved retrieval performance over standard (word based) queries on the same data when a level of 10 percent degradation or worse was achieved. A second method of using n-grams to identify appropriate matching and near matching terms for query expansion which also performed better than using standard queries is also described. This method was less effective than direct n-gram query formulations but can likely be improved with alternative query component weighting schemes and measures of term similarity. Finally, a web based retrieval application using n-gram retrieval of OCR text and display, with query term highlighting, of the source document image is described.
Chapter PDF
Similar content being viewed by others
References
Callan, J.P., Croft, W.B. and Harding, S.M.: The INQUERY Retrieval System. In Proceedings of the 3rd International Conference on Database and Expert Systems Applications (1992) 78–83.
Cavnar, W.: Using an N-Gram-Based Document Representation with a Vector Processing Retrieval Model. In Overview of the Third Text Retrieval Conference (TREC-3), D.K. Harman, Editor (1994) 269–278.
Cohen, D.J.: Highlights: Language and Domain-Independent Automatic Indexing Terms for Abstracting. J. Amer. Soc. Info. Sci. 46 (1995) 162–174.
Croft, W.B., Harding, S.M., Taghva, K. and Borsack, J.: An evaluation of Information Retrieval Accuracy with Simulated OCR Output. Symposium of Document Analysis and Information Retrieval (1994).
Pierce, C. and Nicholas, C.: TELLTALE: Experiments in a Dynamic Hypertext Environment for Degraded and Multilingual Data. J. Amer. Sec. Info. Sci 47 (1996) 263–275.
Rice, S., Kanai, J. and Nartker, T.: An Evaluation of Information Retrieval Accuracy. In UNLV Information Science Research Institute Annual Report (1993) 9–20.
Taghva, K., Borsack, J., Condit, A., Erva, S.: The effects of noisy data on text retrieval. In UNLV Information Science Research Institute Annual Report (1993) 71–80.
Taylor, S.L., Lipshutz, M., Dahl. D.A. and Weir, C.: An Intelligent Document Understanding System. In Second International Conference on Document Analysis and Recognition (1993) 107–220.
Turtle, H. and Croft, W.B.: Evaluation of an Inference Network-Based Retrieval Model. ACM Trans. on Info. Sys. 9 (1991) 187–222.
Ukkonen, E.: Approximate String-Matching with Q-grams and Maximal Matches. Theor. Comp. Sci. 92 (1992) 191–211.
Weir, C., Taylor, S.L., Harding, S.M. and Croft, W.B.: The Skeleton Document Image Retrieval System. In Symposium on Document Image Understanding Technologies (1997).
Zamora, A.: Automatic Detection and Correction of Spelling Errors in a Large Data Base. J. Amer. Soc. Info. Sci. 31 (1980) 51–57.
Zobel, J. and Dart, P.: Finding Approximate Matches in Large Lexicons. Soft. Pract. and Exper. 25 (1995) 331–345.
Zobel, J. and Dart, P.: Phonetic String Matching: Lessons from Information Retrieval. In Proceedings 19th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (1996) 166–173.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Harding, S.M., Croft, W.B., Weir, C. (1997). Probabilistic retrieval of OCR degraded text using N-grams. In: Peters, C., Thanos, C. (eds) Research and Advanced Technology for Digital Libraries. ECDL 1997. Lecture Notes in Computer Science, vol 1324. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0026737
Download citation
DOI: https://doi.org/10.1007/BFb0026737
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63554-3
Online ISBN: 978-3-540-69597-4
eBook Packages: Springer Book Archive
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.