A Connectionist Language Generator
Connectionism has been gaining ground as a psychological modelling technique annd shows great promise as a way to build fast, robust systems to perform intelligent tasks. Connectionism contrasts with the older tradition, that of explaining intelligent behaviour in terms of the manipulation of complex symbol structures.
What people are saying - Write a review
We haven't found any reviews in the usual places.
Lexical Knowledge and Word Choice
Syntactic Knowledge and Its Use
Representing and living Relational Information
Details of FIG
Miscellany Regarding Connectionism
Human Language Production
A Model for Natural Translation
activation flow activation levels adjective algorithms appropriate artificial intelligence big peach cat noun cat thingn cat-collector Chapter closed-class words cnoun Cognitive computation concepts connectionism connectionist constituents Construction Grammar constructions count noun cursor defec defel defes defew defjw defn defr discussed emitted example expresses F Model FIG's Figure Fillmore fromito goals grammar grapheme inf handle highly activated implementation incremental inference input interlingua intermediate representation japanese John language production lexical linguistic Locality Principle machine translation Mary meaning mechanism natural language natural language processing nebors notion output parser parsing part-wise parallelism passive prep-phr preposition problem produce profile space prototype receive activation rel-collector-a relations relevant representation representing requires role Section 5.1 semantic activation semantic update sentence setq simply specific Stemberger subj-pred synergy syntactic categories syntactic structures syntax task thematic role things tion Tsujii tverb utterance verb volitionalr weights word choice world knowledge Yamanashi
Page xi - The research was supported in part by the Defense Advanced Research Projects Agency (DOD) monitored by the Space and Naval Warfare Systems Command under Contract N00039-87-C0251, and in part by the Office of Naval Research under Contracts N00014-87-K-0385 and N00014-87-K-0533.
Page 11 - FIG algorithm is: 1. each node of the input conceptualization is a source of activation 2. activation flows through the network 3. when the network settles, the most highly activated word is selected and emitted 4. activation levels are updated to represent the new current state 5. steps 2 through 4 repeat until all of the input has been conveyed An utterance is simply the result of successive word choices, thus FIG is incremental in a strong sense.