%Aigaion2 BibTeX export from Idiap Publications
%Thursday 21 November 2024 12:12:19 PM

@ARTICLE{vincia04a-art,
         author = {Vinciarelli, Alessandro},
       projects = {Idiap},
          title = {Noisy Text Categorization},
        journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
         volume = {27},
         number = {12},
           year = {2005},
           note = {IDIAP-RR 04-03},
       crossref = {vincia04a},
       abstract = {This work presents categorization experiments performed over noisy texts. By noisy it is meant any text obtained through an extraction process (affected by errors) from media other than digital texts (e.g. transcriptions of speech recordings extracted with a recognition system). The performance of a categorization system over the clean and noisy (Word Error Rate between 10 and 50 percent) versions of the same documents is compared. The noisy texts are obtained through Handwriting Recognition and simulation of Optical Character Recognition. The results show that the performance loss is acceptable and it is especially low for Recall values lower than 60 percent. New measures of the extraction process performance, allowing a better explanation of the categorization results, are proposed.},
            pdf = {https://publications.idiap.ch/attachments/reports/2004/rr04-03.pdf},
     postscript = {ftp://ftp.idiap.ch/pub/reports/2004/rr04-03.ps.gz},
ipdmembership={vision},
}



crossreferenced publications: 
@TECHREPORT{vincia04a,
         author = {Vinciarelli, Alessandro},
       projects = {Idiap},
          title = {Noisy Text Categorization},
           type = {Idiap-RR},
         number = {Idiap-RR-03-2004},
           year = {2004},
    institution = {IDIAP},
           note = {accepted for publication by IEEE Transactions on Pattern Analysis and Machine Intelligence},
       abstract = {This work presents categorization experiments performed over noisy texts. By noisy it is meant any text obtained through an extraction process (affected by errors) from media other than digital texts (e.g. transcriptions of speech recordings extracted with a recognition system). The performance of a categorization system over the clean and noisy (Word Error Rate between 10 and 50 percent) versions of the same documents is compared. The noisy texts are obtained through Handwriting Recognition and simulation of Optical Character Recognition. The results show that the performance loss is acceptable and it is especially low for Recall values lower than 60 percent. New measures of the extraction process performance, allowing a better explanation of the categorization results, are proposed.},
            pdf = {https://publications.idiap.ch/attachments/reports/2004/rr04-03.pdf},
     postscript = {ftp://ftp.idiap.ch/pub/reports/2004/rr04-03.ps.gz},
ipdmembership={vision},
}