Document Classification and Interpretation through the Inference of Logic-Based Models
We present a methodology for document processing that exploits logic-based machine learning techniques. Our claim is that information capture and indexing can profit by the identification of the document class and of specific function of its single layout components. Indeed, the application of incremental and multistrategy machine learning techniques, rather than the classic ones, allows for an efficient solution to the problem of information capture.
KeywordsPredictive Accuracy Digital Library Document Image Document Processing Layout Structure
Unable to display preview. Download preview PDF.
- O. Altamura, F. Esposito, and D. Malerba. Transforming paper documents into XML format with WISDOM++. International Journal on Document Analysis and Recognition, 2001. To appear.Google Scholar
- H. Brocks, U. Thiel, A. Stein, and A. Dirsch-Weigand. Customizable retrieval functions based on user tasks in the cultural heritage domain. In this book.Google Scholar
- E.A. Fox. How to make intelligent digital libraries. In Z.W. Raś and M. Zemankova, editors, Proceedings of the 8th International Symposium on Methodologies for Intelligent Systems, volume 869 of LNAI, pages 27–38. Springer, 1994.Google Scholar
- X. Li and P. Ng. A document classification and extraction system with learning ability. In Proceedings of the 5th International Conference on Document Analysis and Recognition, pages 197–200, 1999.Google Scholar
- F. Sebastiani. Machine learning in automated text categorization. Technical Report Technical Report IEI:B4-31-12-99, CNR-IEI, Pisa, Italy, 1999. Rev. 2001.Google Scholar
- G. Semeraro, F. Esposito, D. Malerba, N. Fanizzi, and S. Ferilli. Machine learning + on-line libraries = IDL. In C. Peters and C. Thanos, editors, Research and Advanced Technology for Digital Libraries. First European Conference-ECDL97, volume 1324 of LNCS, pages 195–214. Springer, 1997.CrossRefGoogle Scholar