Now showing items 1-3 of 3

    • Evaluation of dynamic bayesian network models for entity name transliteration 

      Nabende, Peter (BeNeLearn 09, 2009-05-17)
      This paper proposes an evaluation of DBN models so as to identify DBN configurations that can improve machine transliteration accuracy
    • Translating transliterations 

      Tiedemann, Jörg; Nabende, Peter (Fountain Publishers, Kampala., 2009-08-03)
      Translating new entity names is important for improving performance in Natural Language Processing (NLP) applications such as Machine Translation (MT) and Cross Language Information Retrieval (CLIR). Usually, transliteration ...
    • Transliteration system using pair HMM with weighted FSTs 

      Nabende, Peter (World Scientific Publishing Co Pte Ltd, 2009-08-07)
      This paper presents a transliteration system based on pair Hidden Markov Model (pair HMM) training and weighted Finite State Transducer (WFST)techniques. Parameters used by WFSTs for transliteration generation are learned ...