A Computational Model of Natural Language Communication: by Roland R. Hausser

By Roland R. Hausser

The perfect of utilizing human language to regulate machines calls for a realistic concept of average language verbal exchange that comes with grammatical research of language indicators, plus a version of the cognitive agent, with interfaces for attractiveness and motion, an inner database, and an set of rules for analyzing content material out and in. This e-book deals a sensible framework for theoretical research of ordinary language verbal exchange and for functional functions of average language processing.

Show description

Read Online or Download A Computational Model of Natural Language Communication: Interpretation, Inference, and Production in Database Semantics PDF

Similar linguistics books

English Skills with Readings

The hot 7th version of English abilities with Readings gains John Langan's trademark crystal-clear causes, with his variety of motivating actions and writing assignments that toughen the 4 bases of potent writing: harmony, aid, coherence, and sentence talents. This version of the e-book provides quite a few intriguing new positive factors to John Langan's confirmed technique, and reinstates much-requested fabrics from past variations.

Intermediate Cherokee Language Program

A chain of 5 modules - each one whole with a 1 hour audio tape and workbook and on hand individually. those modules hide the Cherokee verb constitution, together with discussions at the Set A and Set B pronouns,verb conjugations, subject-object prefixes, translation notes and routines, the way to inform time in Cherokee, cultural notes, recipes, legends and poetry by way of Robert Conley.

Additional info for A Computational Model of Natural Language Communication: Interpretation, Inference, and Production in Database Semantics

Example text

4 T HE CYCLE OF NATURAL LANGUAGE COMMUNICATION hearer mode recognition speaker mode LAŦhear LAŦthink LAŦthink sign context proplets action language proplets context proplets language proplets LAŦspeak The schema shows two agents, one in the hearer mode and the other in the speaker mode. Each agent contains its own Word Bank with a context and a language section as well as external interfaces for recognition and action. , LA-hear. 1. The sequencing of word forms during production originates in the time-linear navigation through the set of context proplets in the speaker’s Word Bank, powered by LA-think and indicated as a zig-zag line in the agent on the right.

The ISO standard CD 2460-1 defines a feature structure as a set of feature–value pairs. This is in contradistinction to the use of recursive feature structures within Nativism, for example, in LFG (Bresnan 1982) , GPSG (Gazdar et al. 1985), or HPSG (Pollard and Sag 1994). 5. 36 3. , attribute–value pairs which do not contain feature structures as values. For example, the functor–argument relation between the argument dog and the functor bark is coded by the attribute fnc (functor) of the dog proplet having the value bark, and the attribute arg (argument) of the bark proplet having the value dog.

The rules of LA-think resemble those of LA-hear (cf. 3). The function of the LA-think rules, however, is to power the navigation by retrieving a successor proplet for any given proplet in the Word Bank. 2 (explanations in italics). 1 E XAMPLE OF AN LA-think RULE APPLICATION rule name rule level V_N_V: ss-pattern verb: β arg: X α Y prn: k ⎡ proplet level Cf. FoCL, Part II. } 44 3. Data Structure and Algorithm The ss-pattern of the rule matches an activated proplet in the Word Bank, vertically binding the variable β to the value know, α to Julia or John, and k to 22.

Download PDF sample

Rated 4.97 of 5 – based on 43 votes