Machine translation: past, present, future
Machine translation: past, present, future
Parallel distributed processing: explorations in the microstructure of cognition, vol. 1
Self-organized formation of topologically correct feature maps
Neurocomputing: foundations of research
Neurocomputing: foundations of research
Localized versus distributed representations
The handbook of brain theory and neural networks
Shake-and-bake machine translation
COLING '92 Proceedings of the 14th conference on Computational linguistics - Volume 2
Language and Machines: Computers in Translation and Linguistics
Language and Machines: Computers in Translation and Linguistics
CICLing'11 Proceedings of the 12th international conference on Computational linguistics and intelligent text processing - Volume Part II
OpenLogos machine translation: philosophy, model, resources and customization
Machine Translation
| Hi-index | 0.00 |
The conceptual model underlying the Logos System is described, along with original design motivations and objectives. The model is characterized with respect to four fundamental issues, each of which (it is believed) has been addressed in novel ways: (a) how natural language is to be represented; (b) how linguistic knowledge is to be stored, (c) how this knowledge store is to be applied to the input stream, (d) how complexity effects are to be dealt with as the knowledge store grows, year after year, in the quest for fully automatic,high-quality translation (FAHQT). Empirically rather than formally motivated, the Model nevertheless reflects principles derived from assumptions about human sentence processing, which are described. Using the metaphor of a biological neural net, or bionet, with which the Logos Model has parallels, a complex, 57-word sentence is tracked as it proceeds along a pipeline architecture, simulating a hypothesized human model. Limitations of the model are discussed.