Carl Pollard, Ivan Sag. Information-based syntax and semantics. Vol. 1: Fundamentals. CSLI Lecture Notes No. 13, Stanford, CA.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....language for formulating LP statements, i.e. binary LP constraints. For computational purposes such constraints are compiled into the type definitions for grammatical categories. With respect to the ordering domain, our LP constraints differ from the LP constraints commonly assumed in HPSG (Pollard Sag 1987) in that they 201 apply to nonsibling constituents in head domains. While LP constraints control the order of nodes that are not siblings, information is accumulated in trees in such a way that it is always possible to detect a violation of an LP constraint locally by checking sibling nodes. ....

....of encoding LP constraints works not only for verbal domains, but for any projection of a lexical head. The order of quantifiers and adjectives in a noun phrase can be described by LP constraints. INTEGRATION INTO HPSG In this section, our encoding of LP constraints is incorporated into HPSG (Pollard Sag 1987). We deviate from the standard HPSG grammar in the following respects: The features mentioned above for the encoding of LP constraints are added. Only binary branching grammar rules are used. Two new principles for handling LP constraints are added to the grammar. Further we shall assume a ....

Carl Pollard, Ivan Sag. Information-based syntax and semantics. Vol. 1: Fundamentals. CSLI Lecture Notes No. 13, Stanford, CA.

.... formalisms, unification is the meet operation on the meet semi lattice formed by partially ordering the set of feature structures by a subsumption relation [Shieber 86] Following ideas presented by [AYt Kaci 84] and introduced, for example, in the unification based formalism underlying HPSG [Pollard and Sag 87] first order unification is extended to the sorted case using an order sorted signature instead of a fiat one. In most existing implementations, descriptions of feature structure constraints are not directly used as models that satisfy these constraints; instead, they are represented by ....

....The for malism of TFS is based on the notion of inher itance and sets of constraints that categories of the sort signature must satisfy. The formalism supports to express directly principles and generalizations of linguistic theories as they are formulated for example in the framework of HPSG [Pollard and Sag 87] The efficiency of the LIC ap proach has been tested and compared with Wroblewski s method on a sample grammar of HPSG using a few test sentences for parsing and generation. The overall processing time is reduced by 60 70 of the original processing tirhe. See [Emele 91] for further ....

Carl Pollard and Ivan Sag. Information-Based Syntax and Semantics, Volume I. CSLI Lecture Notes No 13. Chicago University Press, Chicago, 1987.

....language for formulating LP statements, i.e. binary LP constraints. For computational purposes such constraints are compiled into the type definitions for grammatical categories. With respect to the ordering domain, our LP constraints differ from the LP constraints commonly assumed in HPSG (Pollard Sag 1987) in that they apply to nonsibling constituents in head domains. While LP constraints control the order of nodes that are not siblings, information is accumulated in trees in such a way that it is always possible to detect a violation of an LP constraint locally by checking sibling nodes. This ....

....of encoding LP constraints works not only for verbal domains, but for any projection of a lexical head. The order of quantifiers and adjectives in a noun phrase can be described by LP constraints. INTEGRATION INTO HPSG In this section, our encoding of LP constraints is incorporated into HPSG (Pollard Sag 1987). We deviate from the standard HPSG grammar in the following respects: The features mentioned above for the encoding of LP constraints are added. Only binary branching grammar rules are used. Two new principles for handling LP constraints are added to the grammar. Further we shall assume a ....

Carl Pollard, Ivan Sag. Information-based syntax and semantics. Vol. 1: Fundamentals. CSLI Lecture Notes No. 13, Stanford, CA.

.... The past decade has witnessed the emergence and widespread acceptance of declarative grammar formalisms, some well known exemplars being Definite Clause Grammar [Pereira and Warren 1980] Generalized Phrase Structure Grammar [Gazdar et al. 1985] and Head driven Phrase Structure Grammar [Pollard and Sag 1987]. In contradistinction to their more procedural predecessors like Transformational Grammar [Chomsky 1965] and Augmented Transition Networks [Woods 1970] today s declarative grammar formalisms do not prescribe an order in which the possible operations on the grammar are to be carried out 1 . ....

....only rules are chosen which are licensed by the words in the input string, the directed parsing methods [Kay 1980, Wirn 1987] which combine the merits of top down and 2 This is the case for rule based grammars. For principle based grammars like Government Binding Theory [Chomsky 1981] or HPSG [Pollard and Sag 1987], a parsing strategy might specify which principle to apply first and to which linguistic object. bottom up parsing. Head driven parsing also falls into this category. b) Best first parsing (heuristically guided search) A parsing strategy is a heuristically guided search strategy, Kay 1980] ....

Carl Pollard and Ivan Sag. Informationbased Syntax and Semantics. Volume 1: Fundamentals. CSLI Lecture Notes No. 13, Stanford, CA, 1987.

....represenconstruction enrichment text text Figure 5: Interface between sentence and text representation transparent. This can be achieved for example by realizing the sentence representation as one constituent of the text representation. Typed) feature structures like in HPSG (cf. Pollard Sag 87] offer a possibility for the integration of sentence representations into a textual one. An example for the integration of the sentence representations in the text representation is the KBMT system, which is an Interlingua based MT system (cf. Nirenburg et al. 92] On the interlingual level ....

C. Pollard, I. Sag, Information-Based Syntax and Semantics, Volume 1: Fundamentals, CSLI Lecture Notes No. 13, Stanford 1987

....cosma installation that are relevant to the application task at hand. The DISCO system is a reusable NL core machine that relates NL sentences to surfacelevel semantic expressions and vice versa. Analysis and generation processes are based on a declarative and constraint based HPSG style grammar [Pollard and Sag 1987]. This allows for a highly modular, although strongly interconnected, representation of linguistic knowledge [Netter 1993] Although the analytical depth of the core machinery was restricted in order to preserve its domain and task independency, the uniform constraint formalism adopted [Krieger ....

Carl J. Pollard and Ivan A. Sag. Information-Based Syntax and Semantics. Vol. 1: Fundamentals. Chicago (Univ. of Chicago Press), 1987. Center for the Studies of Language and Information, CSLI Lecture Note No. 13.

.... formalisms, unification is the meet operation on the meet semi lattice formed by partially ordering the set of feature structures by a subsumption relation [Shieber 86] Following ideas presented by [Ait Kaci 84] and introduced, for example, in the unification based formalism underlying HPSG [Pollard and Sag 87] first order unification is extended to the sorted case using an order sorted signature instead of a flat one. In most existing implementations, descriptions of feature structure constraints are not directly used as models that satisfy these constraints; instead, they are represented by directed ....

....The formalism of TFS is based on the notion of inheritance and sets of constraints that categories of the sort signature must satisfy. The formalism supports to express directly principles and generalizations of linguistic theories as they are formulated for example in the framework of HPSG [Pollard and Sag 87] The efficiency of the LIC approach has been tested and compared with Wroblewski s method on a sample grammar of HPSG using a few test sentences for parsing and generation. The overall processing time is reduced by 60 70 of the original processing time. See [Emele 91] for further ....

Carl Pollard and Ivan Sag. Information-Based Syntax and Semantics, Volume I. CSLI Lecture Notes No 13. Chicago University Press, Chicago, 1987.

....of the sparse argument list. The term f( b( c(d, e) would thus be represented as the term [2,4 [1,3] and graphically as: 2 4 3 1 3.3. Attribute Value Matrices. Attribute Value Matrices (AVMs) or Feature Structures, are a tool used in several computational linguistic systems (e.g. [4]) Some implementations of AVMs using ordinary terms require prior knowledge of all the attributes an AVM may contain in order to compile appropriate terms (e.g. 5] A simple modification to our scheme, allowing atomic, rather than numeric, indices (for the attributes) and omitting functor names ....

C. Pollard and I. Sag. Information-Based Syntax and Semantics, Volume 1: Fundamentals. CSLI Lecture Notes No. 13, Stanford, CA. 1987.

....provides an environment in which new techniques can be integrated. 1 Introduction Taxonomies, or partial orders, are widespread throughout many disciplines. In computing science, they have importance in databases [1] programming languages [3, 4] operating systems [15] computational linguistics [16, 21], knowledge representation [22] and machine learning [17] As systems scale in size, the need to store and manipulate these taxonomies efficiently becomes increasingly evident. Taxonomic encoding addresses the need to expeditiously perform such operations as greatest lower bounds and comparability ....

C. Pollard and I. Sag. Information-Based Syntax and Semantics. CSLI Lecture Notes No. 13, Stanford, CA, 1987.

....We suggest some ways to limit the expressiveness of sort reasoning to gain polynomial time operation. 1 Introduction Partial orders have been used to represent hierarchically related information, from organization in ecology and biology [6, 14] to representations of lexicons in linguistics [15], but they have been most extensively utilized in sorted logic (e.g. 3, 4] and AI (e.g. description logics [17] machine learning [13] computational linguistics [12] conceptual structures [7, 9] and logic programming [2] The mathematical basis of partial orders has been exploited in ....

C. Pollard and I. Sag. Information-Based Syntax and Semantics. CSLI Lecture Notes No. 13. Stanford, CA, 1987.

....is more basic. Osgood and Bock say that the generative linguists have not agreed on which of these is more basic. Their paper was published in 1977, and the indications I have are that modern linguistic analysis agrees with Osgood s analysis, as I could discern from the examples of Pollard and Sag [63] and Falk [7] On the other hand, with regard to active vs passive voice, both linguists and psycholinguists concur that the active voice is the basic construction. Here, the natural salience of agent over object is clearly available for introspection. In the genitive pair (from [58] a) The ....

....as (a) Red follows blue (b) Blue is followed by red He found that the area assigned to the earlier colour in the sentence was typically larger than that for the later colour. Linguistics Research on linear precedence in linguistics, most notably those of Uszkoreit ( 74] 75] and Pollard and Sag [63], recognizes the presence of competing CHAPTER 5. SALIENCE INTERACTIONS IN PHRASE CHOICE 73 determinants of linear precedence. The competition is discussed in terms of conflicts among ordering rules in the grammar. The elements ordered in these rules are syntactic constituents like NP, PP, etc. ....

[Article contains additional citation context not shown here]

C. Pollard and I.A. Sag. Information-Based Syntax and Semantics, Volume 1: Fundamentals. CSLI, 1987. CSLI Lecture Notes No. 13.

....Some modifications of standard HPSG We assume some familiarity with HPSG as a syntactic theory. The semantics component of HPSG is based on Situation Semantics, the constructs of which are represented as feature structures. The basics of HPSG s semantic description are described in chapter 4 of [Pollard Sag 1987]. In essence, we follow this approach, and we indicate hereafter which modifications have been necessary. In HPSG, the semantic description of a proposition is given in parameterized states of affairs, psoas. A psoa is a constellation of relations, roles and polarities 6 . For each relation, ....

....shows the content value of the nominal projection (the) young man that snores. The restrictions constrain the set of entities by which the parameter can be interpreted. 5 For an overview of TFS, see [Emele 1993] and [Zajac 1992] 6 Polarity is not kept track of in [Pollard Sag, in press] and [Pollard Sag 1987], because there is no analysis of negation so far. 11) 2 6 6 6 6 6 6 6 6 4 para 1 2 6 4 pers third num sing gend masc 3 7 5 para restr ( reln man inst 1 # ; reln young inst 1 # ; reln snore snorer 1 #) 3 7 7 7 7 7 7 7 7 5 nom obj By means of the Semantics ....

[Article contains additional citation context not shown here]

Carl Pollard and Ivan Sag: An Information-based Syntax and Semantics, vol. I, CSLI lecture notes no. 13, 1987.

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