Paper: Mitigation of Data Sparsity in Classifier-Based Translation

ACL ID W08-1501
Title Mitigation of Data Sparsity in Classifier-Based Translation
Venue Workshop on Cognitive Aspects of the Lexicon
Session
Year 2008
Authors

The concept classifier has been used as a translation unit in speech-to-speech trans- lation systems. However, the sparsity of the training data is the bottle neck of its effectiveness. Here, a new method based on using a statistical machine translation system has been introduced to mitigate the effects of data sparsity for training classi- fiers. Also, the effects of the background model which is necessary to compensate the above problem, is investigated. Exper- imental evaluation in the context of cross- lingual doctor-patient interaction applica- tion show the superiority of the proposed method.