Abstract | ||
---|---|---|
The traditional weighting schemes used in text categorization for the vector space model (VSM) cannot exploit information intrinsic to texts obtained through on-line handwriting recognition or any OCR process. Especially, top n (n 1) recognition candidates could not be used without flooding the resulting text with false occurrences of spurious terms. In this paper, an improved weighting scheme for text categorization, that estimates the occurrences of terms from the posterior probabilities of the top n candidates, is proposed. The experimental results show that the categorization performances increase for texts with high error rates. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/ICDAR.2009.137 | ICDAR-1 |
Keywords | Field | DocType |
document image processing,handwritten character recognition,optical character recognition,probability,text analysis,OCR process,handwriting recognition,online handwritten document categorization,posterior probability,recognition candidate,text categorization,vector space model,noisy text categorization,recognition candidates,weighting scheme | Categorization,Weighting,Pattern recognition,Noise measurement,Computer science,Support vector machine,Handwriting recognition,Optical character recognition,Posterior probability,Natural language processing,Artificial intelligence,Vector space model | Conference |
Citations | PageRank | References |
0 | 0.34 | 11 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Sebastián Peña Saldarriaga | 1 | 15 | 4.06 |
Emmanuel Morin | 2 | 42 | 16.13 |
Christian Viard-Gaudin | 3 | 444 | 46.20 |