Summary of the paper

Title Applying a Dynamic Bayesian Network Framework to Transliteration Identification
Authors Peter Nabende
Abstract Identification of transliterations is aimed at enriching multilingual lexiconsand improving performance in various Natural Language Processing (NLP)applications including Cross Language Information Retrieval (CLIR) and MachineTranslation (MT). This paper describes work aimed at using the widely appliedgraphical models approach of ‘Dynamic Bayesian Networks (DBNs) totransliteration identification. The task of estimating transliterationsimilarity is not very different from specific identification tasks where DBNshave been successfully applied; it is also possible to adapt DBN models fromthe other identification domains to the transliteration identification domain.In particular, we investigate the applicability of a DBN framework initiallyproposed by Filali and Bilmes (2005) to learn edit distance estimationparameters for use in pronunciation classification. The DBN framework enablesthe specification of a variety of models representing different factors thatcan affect string similarity estimation. Three DBN models associated with twoof the DBN classes originally specified by Filali and Bilmes (2005) have beentested on an experimental set up of Russian-English transliterationidentification. Two of the DBN models result in high transliterationidentification accuracy and combining the models leads to even much bettertransliteration identification accuracy.
Language Tools, systems, applications
Topics Information Extraction, Information Retrieval, Text mining, Tools, systems, applications
Full paper Applying a Dynamic Bayesian Network Framework to Transliteration Identification
Bibtex @InProceedings{NABENDE10.906,
  author = {Peter Nabende},
  title = {Applying a Dynamic Bayesian Network Framework to Transliteration Identification},
  booktitle = {Proceedings of the Seventh conference on International Language Resources and Evaluation (LREC'10)},
  year = {2010},
  month = {may},
  date = {19-21},
  address = {Valletta, Malta},
  editor = {Nicoletta Calzolari (Conference Chair), Khalid Choukri, Bente Maegaard, Joseph Mariani, Jan Odjik, Stelios Piperidis, Mike Rosner, Daniel Tapias},
  publisher = {European Language Resources Association (ELRA)},
  isbn = {2-9517408-6-7},
  language = {english}
 }
Powered by ELDA © 2010 ELDA/ELRA