A distributed neural network model called SPEC for processing sentences with recursive relative clauses is described. The model is based on sep- arating the tasks of segmenting the input word sequence into clauses, forming the case-role rep- resentations, and keeping track of the recursive embeddings into different modules. The system needs to be trained only with the basic sentence constructs, and it generalizes not only to new in- stances of familiar relative clause structures, but to novel structures as well. SPEC exhibits plausi- ble memory degradation as the depth of the center embeddings increases, its memory is primed by earlier constituents, and its performance is aided by semantic constraints between the constituents. The ability to process structure is largely due to a central executive network that monitors and con- trols the execution of the entire system. This way, in contrast to earlier subsymbolic systems, parsing is modeled as a controlled high-level pro- cess rather than one based on automatic reflex responses.

PDF

In Proceedings of the Twelfth National Conference on Artificial Intelligence, 858-864, January 1994.

@inproceedings{miikkulainen:aaai94, title={Parsing Embedded Clauses with Distributed Neural Networks}, author={Risto Miikkulainen and Dennis Bijwaard}, booktitle={Proceedings of the Twelfth National Conference on Artificial Intelligence}, month={January}, pages={858-864}, url="http://nn.cs.utexas.edu/?bijwaard:aaai94", year={1994} }