P. Saint-Dizier. Advanced Logic Programming for Language Processing. Academic Press, 1994.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....one needs not even know the application. Such an abstraction is not illegal. It represents exactly all functions that ignore their argument (i.e. constant functions) but these functions are seldom of interest. However, they can be found in print where constant functions are not intended (e.g. [51], pp. 263, 265 269) All this shows that programming with Prolog requires new reflexes that are not so well known. 3.2.4 Context in programs We give as an example of context in programs the Prolog code of an unfamiliar implementation of the familiar concatenation relation. type app (list T) ....

....terms offer an equivalent to coreference constraints only. So, the term as rational tree representation can be easily adapted to represent terms with feature terms. It happens that feature terms are also useful for Computational Linguistics: for instance, for implementing Unification Grammars [31, 51]. So, the term as rational tree representation adapted to feature terms allows us to handle several facets of a Natural Language application in a unique framework. An alternative would be to provide Prolog with feature terms; this does not exist yet. 5 A LOGICAL HANDLING OF CONTEXT DCGs can ....

[Article contains additional citation context not shown here]

P. Saint-Dizier. Advanced Logic Programming for Language Processing. Academic Press, 1994.

....one need not even know the application. Such an abstraction is not illegal; it represents exactly all functions that ignore their argument (i.e. constant functions) These functions are seldom of interest, but terms like xnE can be found in print where constant functions are not intended (e.g. [49], pp. 263, 265 269) All this shows that programming with Prolog requires new practices that are not so well known. 6 In that respect, logical variables differ from the syntactic variables used in texts about the calculus like the present article. 7 We recall the j equivalence axiom: x(E x) ....

....Since coreference constraints can emulate rational terms, the term as rational tree representation can be easily adapted to represent terms with feature terms. It happens that feature terms are also useful for Computational Linguistics: for instance, for implementing Unification Grammars [29, 49]. So, the term as rational tree representation adapted to feature terms allows us to handle several facets of a Natural Language application in a unique framework. An alternative would be to provide Prolog with feature terms; this does not exist yet. 5. A LOGICAL HANDLING OF CONTEXT DCGs can ....

[Article contains additional citation context not shown here]

P. Saint-Dizier. Advanced Logic Programming for Language Processing. Academic Press, 1994.

....1.3 Constraint Logic Programming We claim that we can have both declarativity and tractability if we use the CLP framework [8] possibly with disjunctive rules. This idea is not completely new. There are several suggestions on how to exploit CLP for natural language processing. For example, [23] introduces boolean constraints over feature values and finite domain constraints over feature value domains. In this paper, we will introduce our system for implementing HPSG style grammars, namely the constraint based language UBS (UnifikationsBasierte Sprache) 26] which provides typed feature ....

Patrick Saint-Dizier. Advanced Logic Programming for Language Processing. Academic Press, London, 1994.

No context found.

Saint-Dizier P. (1994) Advanced Logic Programming for Language Processing, Academic Press.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC