Using a lattice to enhance adaptation guided retrieval in Example Based Machine Translation
SL Serutla, G. Deon Oosthuizen · Unisa Institutional Repository (University of South Africa) · 2000
Ease of retrieval is of crucial importance in case based reasoning (CBR). In fact, the performance of a CBR system depends on how such a system is able to retrieve relevant cases from memory and adapts them to suit new problem cases. This paper suggests a method of enhancing the retrieval process of relevant cases from memory by employing a lattice, which has potentially good indexing capabilities. It will be shown that the method is well suited for Example Based Machine Translation (EBMT), especially when the input is grammatically incorrect.