Travelled to:
1 × Brazil
1 × Ireland
10 × USA
Collaborated with:
J.S.Coombs J.Borsack T.A.Nartker R.Pereda A.Condit S.Agarwal M.Cartright R.Beckley R.Young S.Poudel S.Malreddy T.Wu J.L.Clark N.Zhou S.E.Lumos
Talks about:
extract (5) ocr (5) document (4) use (4) correct (2) markov (2) inform (2) hidden (2) model (2) text (2)
Person: Kazem Taghva
DBLP: Taghva:Kazem
Facilitated 2 volumes:
Contributed to:
Wrote 13 papers:
- DRR-2014-TaghvaA #fault #identification #web
- Utilizing web data in identification and correction of OCR errors (KT, SA), pp. 902109–6.
- DRR-2013-TaghvaPM #higher-order #markov #modelling
- Post processing with first- and second-order hidden Markov models (KT, SP, SM).
- DRR-2010-PeredaT #precise #semiparsing #using
- Date of birth extraction using precise shallow parsing (RP, KT), pp. 1–10.
- DRR-2009-TaghvaC #analysis #bound #documentation #using
- Document boundary determination using structural and lexical analysis (KT, MAC), pp. 1–10.
- DRR-2007-TaghvaCLBN #documentation #generative
- Title extraction and generation from OCR’d documents (KT, AC, SEL, JB, TAN).
- ICDAR-2007-TaghvaBC #documentation
- Extracting _Carbon Copy_ Names and Organizations from a Heterogeneous Document Collection (KT, RB, JSC), pp. 664–668.
- DRR-2006-TaghvaBCBPN #automation #information management #relational #using
- Automatic redaction of private information using relational information extraction (KT, RB, JSC, JB, RP, TAN).
- DRR-2005-TaghvaCPN #markov #modelling #using
- Address extraction using hidden Markov models (KT, JSC, RP, TAN), pp. 119–126.
- DRR-2004-TaghvaBNCY #proximity
- The impact of running headers and footers on proximity searching (KT, JB, TAN, JSC, RY), pp. 1–5.
- DRR-2003-NartkerTYBC #documentation
- OCR correction based on document level knowledge (TAN, KT, RY, JB, AC), pp. 103–110.
- DRR-2003-TaghvaC #categorisation #question #rule-based
- Do Thesauri enhance rule-based categorization for OCR text? (KT, JSC), pp. 111–119.
- SIGIR-1994-TaghvaBC #information retrieval #probability
- Results of Applying Probabilistic IR to OCR Text (KT, JB, AC), pp. 202–211.
- SEKE-1989-WuCZT #identification #morphism #novel #query #subclass
- A Novel Way 1o Identify IneguaIity Query Subclasses Which possess the Homomorphism Property (TW, JLC, NZ, KT), pp. 158–163.