An Image Understanding System Using Attributed Symbolic Representation and Inexact Graph-Matching
IEEE Transactions on Pattern Analysis and Machine Intelligence
Generating and generalizing models of visual objects
Artificial Intelligence
Classification of newspaper image blocks using texture analysis
Computer Vision, Graphics, and Image Processing
The skew angle of printed documents
Document image analysis
A Comparative Analysis of Methods for Pruning Decision Trees
IEEE Transactions on Pattern Analysis and Machine Intelligence
Document Processing for Automatic Knowledge Acquisition
IEEE Transactions on Knowledge and Data Engineering
Incremental Induction of Decision Trees
Machine Learning
Representing OCRed documents in HTML
ICDAR '97 Proceedings of the 4th International Conference on Document Analysis and Recognition
Handling Continuous Data in Top-Down Induction of First-Order Rules
AI*IA '97 Proceedings of the 5th Congress of the Italian Association for Artificial Intelligence on Advances in Artificial Intelligence
Processing Paper Documents with WISDOM
AI*IA '97 Proceedings of the 5th Congress of the Italian Association for Artificial Intelligence on Advances in Artificial Intelligence
WISDOM++: An Interactive and Adaptive Document Analysis System
ICDAR '99 Proceedings of the Fifth International Conference on Document Analysis and Recognition
A knowledge-based approach to the layout analysis
ICDAR '95 Proceedings of the Third International Conference on Document Analysis and Recognition (Volume 1) - Volume 1
ANASTASIL: hybrid knowledge-based system for document layout analysis
IJCAI'89 Proceedings of the 11th international joint conference on Artificial intelligence - Volume 2
IBM Journal of Research and Development
Adaptive document block segmentation and classification
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
| Hi-index | 0.00 |
WISDOM++ is an intelligent document processing system that transforms a paper document into HTML/XML format. The main design requirement is adaptivity, which is realized through the application of machine learning methods. This paper illustrates the application of symbolic learning algorithms to the first three steps of document processing, namely document analysis, document classification and document understanding. Machine learning issues related to the application are: Efficient incremental induction of decision trees from numeric data, handling of both numeric and symbolic data in first-order rule learning, learning mutually dependent concepts. Experimental results obtained on a set of real-world documents are illustrated and commented.