A maximum entropy approach to natural language processing
Computational Linguistics
The Perceptron Algorithm with Uneven Margins
ICML '02 Proceedings of the Nineteenth International Conference on Machine Learning
Probabilistic Retrieval of OCR Degraded Text Using N-Grams
ECDL '97 Proceedings of the First European Conference on Research and Advanced Technology for Digital Libraries
Offline Recognition of Large Vocabulary Cursive Handwritten Text
ICDAR '03 Proceedings of the Seventh International Conference on Document Analysis and Recognition - Volume 2
Holistic Word Recognition for Handwritten Historical Documents
DIAL '04 Proceedings of the First International Workshop on Document Image Analysis for Libraries (DIAL'04)
The class imbalance problem: A systematic study
Intelligent Data Analysis
Gabor features for offline Arabic handwriting recognition
DAS '10 Proceedings of the 9th IAPR International Workshop on Document Analysis Systems
Hi-index | 0.00 |
This paper investigates different machine learning models to solve the historical handwritten manuscript recognition problem. In particular, we test and compare support vector machines, conditional maximum entropy models and Naive Bayes with kernel density estimates and explore their behaviors and properties when solving this problem. We focus on a whole word problem to avoid having to do character segmentation which is difficult with degraded handwritten documents. Our results on a publicly available standard dataset of 20 pages of George Washington's manuscripts show that Naive Bayes with Gaussian kernel density estimates significantly outperforms the other models and prior work using hidden Markov models on this heavily unbalanced dataset.