| { | |
| "paper_id": "W98-0109", | |
| "header": { | |
| "generated_with": "S2ORC 1.0.0", | |
| "date_generated": "2023-01-19T06:03:29.545594Z" | |
| }, | |
| "title": "An\" Integrated Parser for TFG with Explicit Tree Typing 1. lntroduction", | |
| "authors": [ | |
| { | |
| "first": "Marc", | |
| "middle": [], | |
| "last": "Ca", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "University of Bradford Bradford", | |
| "location": { | |
| "addrLine": "BD7 lDP", | |
| "country": "United Kingdom" | |
| } | |
| }, | |
| "email": "[email protected]" | |
| } | |
| ], | |
| "year": "", | |
| "venue": null, | |
| "identifiers": {}, | |
| "abstract": "http:l/www.eimc.brad.ac. uk/-mcavazza One of\"the main conditions for the development of successful NLP applications is the usability of the syntactic formalisms adopted and the degree to which they facilitate syntax-semantics integration. TAG+ formalisms show a real potential for NLP applications, due to their linguistic descriptive capabilities. However, both the standard formalisms and parsing strategies are often quite complex and cannot easily be used as such for the development of NLP systems. We have .thus investigated a simplified TAG+ fonnalism, which sacrifices some of TAG's descriptive and formal properties for the sake of usability. Titls is especially relevant considering the recent success of empirical approaches to NLP which tend tobe based on very simple techniques and/or discard linguistically-motivated formalisms [Basili et al\" 1996] [Appelt et al., 1993]. We report the implementation of a parser for a simplified TAG+ fonnalism, Tree Furcating Grammars (TFG), which integrales semantic processing, performing both syntactic disambiguation and the construction of a semantic representation for the sentence parsed. The parser has been developed for the purpose of real-time speech understanding of sublanguages (i.e., application-dependent vocabularies of 500-1000 words with specific, sometimes quite simptified, syntactic constructs). TAG+ formalisms were initially investigated because of their potential for syntax-semantics integration (see e.g., Abeille [1994)). We will successively describe the rationale for the TFG formalism, the principles underlying the algorithm used and a first assessment of its perfonnance.", | |
| "pdf_parse": { | |
| "paper_id": "W98-0109", | |
| "_pdf_hash": "", | |
| "abstract": [ | |
| { | |
| "text": "http:l/www.eimc.brad.ac. uk/-mcavazza One of\"the main conditions for the development of successful NLP applications is the usability of the syntactic formalisms adopted and the degree to which they facilitate syntax-semantics integration. TAG+ formalisms show a real potential for NLP applications, due to their linguistic descriptive capabilities. However, both the standard formalisms and parsing strategies are often quite complex and cannot easily be used as such for the development of NLP systems. We have .thus investigated a simplified TAG+ fonnalism, which sacrifices some of TAG's descriptive and formal properties for the sake of usability. Titls is especially relevant considering the recent success of empirical approaches to NLP which tend tobe based on very simple techniques and/or discard linguistically-motivated formalisms [Basili et al\" 1996] [Appelt et al., 1993]. We report the implementation of a parser for a simplified TAG+ fonnalism, Tree Furcating Grammars (TFG), which integrales semantic processing, performing both syntactic disambiguation and the construction of a semantic representation for the sentence parsed. The parser has been developed for the purpose of real-time speech understanding of sublanguages (i.e., application-dependent vocabularies of 500-1000 words with specific, sometimes quite simptified, syntactic constructs). TAG+ formalisms were initially investigated because of their potential for syntax-semantics integration (see e.g., Abeille [1994)). We will successively describe the rationale for the TFG formalism, the principles underlying the algorithm used and a first assessment of its perfonnance.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Abstract", | |
| "sec_num": null | |
| } | |
| ], | |
| "body_text": [ | |
| { | |
| "text": "Tree Furcating Grammars are a lexicalised TAG+ formalism, in which adjunction is replaced by the furcation operation that essentially adds an additional branch to the target node in the initial tree, instead of copying the auxiliary tree under it. The furcation operation was originally introduced in segrnent grammars [De Smedt & Kempen, 1990] . A detailed comparison of furcation and adjunction has been given by Abeille [ 1991] . Though some syntactic phenomena are not properly bandled by furcation, the fact that it introduces modifiers without embedding them into the tree structure is a definite advantage for syntax-semantics integration, and was the rationale for choosing it 1 \u2022 Successive furcations do not increase tree depth and complexity, producing deriv~ trees that retain some properties of dependency trees. These can support the integrated construction of a semantic structure, based on the appropriate association of semantic functions to the tree structures (see below).", | |
| "cite_spans": [ | |
| { | |
| "start": 319, | |
| "end": 344, | |
| "text": "[De Smedt & Kempen, 1990]", | |
| "ref_id": "BIBREF7" | |
| }, | |
| { | |
| "start": 415, | |
| "end": 430, | |
| "text": "Abeille [ 1991]", | |
| "ref_id": "BIBREF0" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The Tree Furcating Grammars (TFG) Formalism", | |
| "sec_num": "2." | |
| }, | |
| { | |
| "text": "We have adapted our tree representations accordingly, by distinguishing between left auxiliary trees (which have a *X root node) 2 and right auxiliary trees (X* root node). The aux.iliary symbol is on the root node, as these trees do not have a foot node. Also, in our implementation trees are explicitly typed as left or right auxiliary (1-aux, r-aux), initial and left or right substituable (l-subst, r-subst). Trees can have multiple types, for instance being both right and left substituable or, in the case of some PP trees, both left auxiliary and right substituable (e.g., fig 2, *V-with-NO).", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The Tree Furcating Grammars (TFG) Formalism", | |
| "sec_num": "2." | |
| }, | |
| { | |
| "text": "Left (resp. right) auxiliary trees are combined through right (resp. left) f\u00f6rcation. Left and right furcations, as described by De Smedt & Kempen [1990] produce \"flat\" structures and in that sense differ from left and right adjunction in Tree Insertion Grammars [Schabes & Waters, 1994] . They tend tobe closer to the operations described by Nasr [1995] for his dependency-based TAG variant. Also, furcations are not allowed to take place at substituable nodes prior to their substitution, but are allowed on au:idliary nodes (as compared with Schabes & Waters (1994] ).", | |
| "cite_spans": [ | |
| { | |
| "start": 129, | |
| "end": 153, | |
| "text": "De Smedt & Kempen [1990]", | |
| "ref_id": "BIBREF7" | |
| }, | |
| { | |
| "start": 263, | |
| "end": 287, | |
| "text": "[Schabes & Waters, 1994]", | |
| "ref_id": "BIBREF18" | |
| }, | |
| { | |
| "start": 343, | |
| "end": 354, | |
| "text": "Nasr [1995]", | |
| "ref_id": "BIBREF11" | |
| }, | |
| { | |
| "start": 545, | |
| "end": 568, | |
| "text": "Schabes & Waters (1994]", | |
| "ref_id": "BIBREF18" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The Tree Furcating Grammars (TFG) Formalism", | |
| "sec_num": "2." | |
| }, | |
| { | |
| "text": "Another goal, which was the result of early experimentation, was to minimise tree traversal operations that can prove computationally expensive. These are minimised due to the representation itself and to the explicit recording of substituable leaves within tree representations. Only the determination of target nodes for furcation still requires tree traversal, but is made easier by the relatively flat structure of the derived trees.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The Tree Furcating Grammars (TFG) Formalism", | |
| "sec_num": "2." | |
| }, | |
| { | |
| "text": "Finally, a set of atomic semantic features, corresponding to the semantic description of the anchor is associated to the root node as well. These semantic features are used for semantic representations as weil as selectional resttictions, in the spirit of preference semantics [Wilks, 1975) . Substituable nodes in initial and some auxiliary trees are associated semantic relations, which also constitute an explicit typing. The definition of these semantic relations can be quite specific, as it deri ves from the specific distributions of the lexicalised trees themsel ves [Cavazza, 1997) .", | |
| "cite_spans": [ | |
| { | |
| "start": 277, | |
| "end": 290, | |
| "text": "[Wilks, 1975)", | |
| "ref_id": "BIBREF19" | |
| }, | |
| { | |
| "start": 575, | |
| "end": 590, | |
| "text": "[Cavazza, 1997)", | |
| "ref_id": "BIBREF5" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The Tree Furcating Grammars (TFG) Formalism", | |
| "sec_num": "2." | |
| }, | |
| { | |
| "text": "Several parsing algorithms have been described for TAG+, including CKY [Vijay-Shanker & Joshi, 1985) and", | |
| "cite_spans": [ | |
| { | |
| "start": 88, | |
| "end": 100, | |
| "text": "Joshi, 1985)", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The Integrated Parsing Algorithm", | |
| "sec_num": "3." | |
| }, | |
| { | |
| "text": "Earley-type parsers and a deterministic parser [Schabes & Vijay-Shanker, 1990 ], which was developed for reasons of efficiency (Schabes & Joshi, 1990) . The Iatter has been recently revisited by Kinyon [1997] , who proposed an improved LR(O) algoritbm. Recently, Nederhof [1998] has described a new LR parsing method and a new recogniser based on Linear lndexed Automata. Specific approaches have also been developed for partial parsing of potentially ungrammatical sentences [Issac, 1994) . Another major source of innovation in parsing has been the many TAG+ variants developed in recent years, such as the \"supertagging\" approach [Joshi & Srinivas, 1994) , dependency formalisms inspired by TAG [Nasr, 1995) and Tree Insertion Grammars [Schabes & Waters, 1994 ].", | |
| "cite_spans": [ | |
| { | |
| "start": 47, | |
| "end": 77, | |
| "text": "[Schabes & Vijay-Shanker, 1990", | |
| "ref_id": "BIBREF17" | |
| }, | |
| { | |
| "start": 127, | |
| "end": 150, | |
| "text": "(Schabes & Joshi, 1990)", | |
| "ref_id": "BIBREF16" | |
| }, | |
| { | |
| "start": 195, | |
| "end": 208, | |
| "text": "Kinyon [1997]", | |
| "ref_id": "BIBREF10" | |
| }, | |
| { | |
| "start": 263, | |
| "end": 278, | |
| "text": "Nederhof [1998]", | |
| "ref_id": "BIBREF12" | |
| }, | |
| { | |
| "start": 476, | |
| "end": 489, | |
| "text": "[Issac, 1994)", | |
| "ref_id": "BIBREF8" | |
| }, | |
| { | |
| "start": 633, | |
| "end": 657, | |
| "text": "[Joshi & Srinivas, 1994)", | |
| "ref_id": "BIBREF9" | |
| }, | |
| { | |
| "start": 698, | |
| "end": 710, | |
| "text": "[Nasr, 1995)", | |
| "ref_id": "BIBREF11" | |
| }, | |
| { | |
| "start": 739, | |
| "end": 762, | |
| "text": "[Schabes & Waters, 1994", | |
| "ref_id": "BIBREF18" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The Integrated Parsing Algorithm", | |
| "sec_num": "3." | |
| }, | |
| { | |
| "text": "Due to the interleaving of syntactic and semantic processing in our system, we have opted for an ad hoc sttategy, which eventually resulted quite similar to the one described by Nasr [1995] . The main idea is to make the syntactic part of the algorithm as simple as possible and to avoid \"hidden\" integration of syntax and semantics through contextual constraints on syntactic operations. Rather, keeping the parsing algorithm elementary would offer more space for experimentation and the integration of semantic processing.", | |
| "cite_spans": [ | |
| { | |
| "start": 178, | |
| "end": 189, | |
| "text": "Nasr [1995]", | |
| "ref_id": "BIBREF11" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The Integrated Parsing Algorithm", | |
| "sec_num": "3." | |
| }, | |
| { | |
| "text": "The first step, which corresponds to a lexical filtering of the granunar, consists in generating all the possible sei of trees (often termedforests) compatible with tbe input string. This step is very similar to the construction of a pushdown stack for !rees as described in [Nasr, 1995] . The parsing algoritbm coasists in scanning the forest lefMo-right and determining possible tree fusions from the explicit typing of the !rees considered. The process is iterated until the forest is reduced to a single tree or no further operations are possible [Cavazza & Constant, 1996] . All the forests not reduced to a single tree are discarded as unsucce.ssful parses. Adjacent ttees in a forest are considered for a possible fusion on a pairwise basis. From their explicit categories, the corresponding operation is given by a compatibility table. This table specifies the nature of the operation (substitution, furcation, or nil) as a function of the types of the adjacent ttees. However, succe.ssful Operations also depend on the existence of an appropriate target node as well as semantic compatibility (when applicable). In that sense, tree operations are not fully determined by the compatibility table. The target node for substitution is directly recorded in the representation for substituable trees, while target node for furcation is dynamically computed as being the rightmost/leftmost compatible nodc, including nodes intemal to the tree. The forest is scanned Jeft to right without look-ahead and the \"cursor\" backtracks one position after a successful fusion has been completed. The forest may have to be scanned severa.I times, due to the conjunction of a strict left-toright scanning with the restrictions imposed on tree operations. For instance, in the parsing of the forest on fig. 2 , the first pass essentially assembles the nominal descriptions through furcation, and substitution at the NI node takes place at the socond pass only.", | |
| "cite_spans": [ | |
| { | |
| "start": 275, | |
| "end": 287, | |
| "text": "[Nasr, 1995]", | |
| "ref_id": "BIBREF11" | |
| }, | |
| { | |
| "start": 551, | |
| "end": 577, | |
| "text": "[Cavazza & Constant, 1996]", | |
| "ref_id": "BIBREF4" | |
| } | |
| ], | |
| "ref_spans": [ | |
| { | |
| "start": 1792, | |
| "end": 1798, | |
| "text": "fig. 2", | |
| "ref_id": null | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "The Integrated Parsing Algorithm", | |
| "sec_num": "3." | |
| }, | |
| { | |
| "text": "Additional heuristics arc used as a declarative control strategy. For instance, whenever a PP tree is both of type 1-aux and r-subst (like e.g\" *V-with-NO), substitution has to be perfonned firsl, thus enabling correct semantic feature propagation, which will be subsequently needed for selectional restriction at (right) furcation time. This can be achieved by attributing precedence to some types; as a result some operations are postponed until proper conditions are met. lt should be noted that PP attachments are a major requirement for the processing of definite descriptions, spatial expressions and instrumental actions, which constitute a significant fraction of the requirements for speech-based multimedia applications.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The Integrated Parsing Algorithm", | |
| "sec_num": "3." | |
| }, | |
| { | |
| "text": "1broughout parsing, there is a f\u00fcll integration of semantic processing\\ which consists both in semantic features propagation and establishment of semantic/functional links for actants and various modifiers. Semantic features for a lexical entry are associated to the tree root and are transferred through furcation operations to the root node of the target tree ( fig. 1 and 2) . This ensures proper propagation of semantic features to constitute complete semantic frames.", | |
| "cite_spans": [], | |
| "ref_spans": [ | |
| { | |
| "start": 364, | |
| "end": 377, | |
| "text": "fig. 1 and 2)", | |
| "ref_id": "FIGREF0" | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "The Integrated Parsing Algorithm", | |
| "sec_num": "3." | |
| }, | |
| { | |
| "text": "Furcation is responsible for semantic aggregation, while substitution establishes semantic relations between meaning units, essentially through the structure of initial trees of root S. However, furcation can also result in the establishment of semantic relations, for instance instrumental cases, as with *V-with-NO trees. Figures 1 and 2 illustrate two different cases of selectional restriction, implementing the PP-atcachrnent rules described above 4", | |
| "cite_spans": [], | |
| "ref_spans": [ | |
| { | |
| "start": 324, | |
| "end": 339, | |
| "text": "Figures 1 and 2", | |
| "ref_id": "FIGREF0" | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "The Integrated Parsing Algorithm", | |
| "sec_num": "3." | |
| }, | |
| { | |
| "text": "Tue system is implemented in Common LISP and runs on a SGI 02 workstation with an RlOOOO processor at 150 MHz. Processing of a single forest corresponding to a 10-15 word sentence is regularly carried in 10-20 ms CPU time.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Results", | |
| "sec_num": "4." | |
| }, | |
| { | |
| "text": "Tue important point is that, even wben parsing several forests for a sentence, the user time remains below 200 ms.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Results", | |
| "sec_num": "4." | |
| }, | |
| { | |
| "text": "Though thls was measured with small vocabularies (typically less than 300 words), it is expected to remain roughly uncbanged with the target application vocabulary being approx. 500 words in size. Tue reason is that global response times depend on the number of forests to parse, which is a function of the trees/word ratio. This ratio tends to remain stable withln smaJl sublanguages and is certainly much smaller than the generic ratio of 7 mentioned in [Schabes & Waters, 1994] . It is interesting to compare these results to the requirements proposed by Goerz and Kessler [1994] for anytime algorithms to be used in speecb understanding. They give Result Production Granularity (RPG) values in the range of 10-100 ms, which means that in most cases our parser, developed for similar applications, could fit into that range.", | |
| "cite_spans": [ | |
| { | |
| "start": 456, | |
| "end": 480, | |
| "text": "[Schabes & Waters, 1994]", | |
| "ref_id": "BIBREF18" | |
| }, | |
| { | |
| "start": 558, | |
| "end": 582, | |
| "text": "Goerz and Kessler [1994]", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Results", | |
| "sec_num": "4." | |
| }, | |
| { | |
| "text": "[:act :openJ", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Results", | |
| "sec_num": "4." | |
| }, | |
| { | |
| "text": "[:artifactJ p N* .... ... \"' 1 \" ... V NO.J.. det N 1 1 1 open the door [:omament] v. .. .... *N \".\" .. [:omament] 1 :\u2022\u2022\u2022\u2022\u2022\u2022\u2022\u2022\u2022\u2022 N \u2022\u2022\u2022\u2022N PP~T l ~P Noi dl'", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Results", | |
| "sec_num": "4." | |
| }, | |
| { | |
| "text": "with the skull the skull key Fig. 2 . Semantic propagation through substitution (''NO\" node ofthe PP group) enables selection of right furcation on \"V\", because of compatibility between \"telic\" features and feature precedence rules.", | |
| "cite_spans": [], | |
| "ref_spans": [ | |
| { | |
| "start": 29, | |
| "end": 35, | |
| "text": "Fig. 2", | |
| "ref_id": null | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "Results", | |
| "sec_num": "4." | |
| }, | |
| { | |
| "text": "\" These refer to situations enoountered in the popular \"DOOM\" video game (trademark of ID Software).", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Results", | |
| "sec_num": "4." | |
| }, | |
| { | |
| "text": "We do not make a direct use of the properties of the derived lree, like dominance relations.2 With X in {P, N, V, A}.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "In that sense, our implementation would fall under the ''Parallel\" + \"Generate-and-Test\" paradigm for Syntax-Semantics integration[Dahl et al\" 1992).", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "", | |
| "sec_num": null | |
| } | |
| ], | |
| "back_matter": [], | |
| "bib_entries": { | |
| "BIBREF0": { | |
| "ref_id": "b0", | |
| "title": "Une grammaire lexicalisw d'arbres adjoints pour l; francais: application a l'analyse automatique", | |
| "authors": [ | |
| { | |
| "first": "A", | |
| "middle": [], | |
| "last": "Abeille", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1991, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Abeille, A\" 1991. Une grammaire lexicalisw d'arbres adjoints pour l; francais: application a l'analyse automatique. These de Doctorat de l'Universite Paris 7.", | |
| "links": null | |
| }, | |
| "BIBREF1": { | |
| "ref_id": "b1", | |
| "title": "Syntaxe et Semantique: Interactions dans une grammaire d'unification", | |
| "authors": [ | |
| { | |
| "first": "A", | |
| "middle": [], | |
| "last": "Abeille", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1994, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Abeille, A., 1994. Syntaxe et Semantique: Interactions dans une grammaire d'unification. In: F. Rastier, M. Cavazza, A. Abeille, Semantique pour l'Analyse. Paris, Masson. (English lranslation to appear, CSU, University of Stanford Press).", | |
| "links": null | |
| }, | |
| "BIBREF2": { | |
| "ref_id": "b2", | |
| "title": "FASTUS: A Finite-state Processor for lnfonnation Extraction from Real-World Text", | |
| "authors": [ | |
| { | |
| "first": "D", | |
| "middle": [ | |
| "E" | |
| ], | |
| "last": "Appell", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "J", | |
| "middle": [ | |
| "R" | |
| ], | |
| "last": "Hobbs", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "", | |
| "middle": [], | |
| "last": "Bear", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "D", | |
| "middle": [], | |
| "last": "Israel", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "M", | |
| "middle": [], | |
| "last": "Tyson", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1993, | |
| "venue": "Proceedings of the IJCA1'93 Conference", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Appell, D. E., Hobbs, J. R., Bear, J\" Israel, D. & Tyson M. (1993). FASTUS: A Finite-state Processor for lnfonnation Extraction from Real-World Text. In Proceedings of the IJCA1'93 Conference.", | |
| "links": null | |
| }, | |
| "BIBREF3": { | |
| "ref_id": "b3", | |
| "title": "An Empirical Symbolic Approach to Natural Language Processing", | |
| "authors": [ | |
| { | |
| "first": "R", | |
| "middle": [], | |
| "last": "Basili", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "M", | |
| "middle": [ | |
| "T" | |
| ], | |
| "last": "Pazienza", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "P", | |
| "middle": [], | |
| "last": "Velarcli", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1996, | |
| "venue": "Artijicial Intelligence", | |
| "volume": "85", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Basili, R., M.T. Pazienza, & Velarcli, P., 1996. An Empirical Symbolic Approach to Natural Language Processing, Artijicial Intelligence, vol. 85.", | |
| "links": null | |
| }, | |
| "BIBREF4": { | |
| "ref_id": "b4", | |
| "title": "Le Traitement Automatique du Langage Naturei et ses Applications", | |
| "authors": [ | |
| { | |
| "first": "M", | |
| "middle": [], | |
| "last": "Cavazza", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "P", | |
| "middle": [], | |
| "last": "Constant", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1996, | |
| "venue": "Nouvelles Interfaces Homme-Machine", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Cavazza, M. & Constant, P., 1996. Le Traitement Automatique du Langage Naturei et ses Applications. In: J. Caelen (Ed.), Nouvelles Interfaces Homme-Machine, Paris: Tee & Doc.", | |
| "links": null | |
| }, | |
| "BIBREF5": { | |
| "ref_id": "b5", | |
| "title": "Semiotique textuelle et contenu linguistique. Intellectica", | |
| "authors": [ | |
| { | |
| "first": "M", | |
| "middle": [], | |
| "last": "Cavazza", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1997, | |
| "venue": "", | |
| "volume": "23", | |
| "issue": "", | |
| "pages": "53--78", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Cavazza, M., 1997. Semiotique textuelle et contenu linguistique. Intellectica, vol. 23, pp. 53-78.", | |
| "links": null | |
| }, | |
| "BIBREF6": { | |
| "ref_id": "b6", | |
| "title": "Integration of Syntax and Semantics in the Pundit System", | |
| "authors": [ | |
| { | |
| "first": "D\"", | |
| "middle": [], | |
| "last": "Dahl", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "C", | |
| "middle": [], | |
| "last": "Weir", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "L", | |
| "middle": [ | |
| "M" | |
| ], | |
| "last": "Norton", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "M", | |
| "middle": [], | |
| "last": "Linebarger", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1992, | |
| "venue": "Proceedings of the ANLP'92 Workshop on Fully-lmplemented NLP Systems, IBM Technical Report", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Dahl, D\" Weir, C., Norton, L.M. & Linebarger, M., 1992. Integration of Syntax and Semantics in the Pundit System. Proceedings of the ANLP'92 Workshop on Fully-lmplemented NLP Systems, IBM Technical Report, TR-80.92-033.", | |
| "links": null | |
| }, | |
| "BIBREF7": { | |
| "ref_id": "b7", | |
| "title": "Segment Granunars: a Formalism for Incremental Sentence Generation", | |
| "authors": [ | |
| { | |
| "first": "K", | |
| "middle": [], | |
| "last": "De Smedt", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "G", | |
| "middle": [], | |
| "last": "Kempen", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1990, | |
| "venue": "Natural lmiguage Generation and Computational I.inguistics, Dordrecht, Kluwer. Goerz, G. & Kesseler, M\" 1994. Anytime Algorithms for Speech Parsing? Proceedings of COUNG '94", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "De Smedt, K. & Kempen, G\" 1990. Segment Granunars: a Formalism for Incremental Sentence Generation. In: C. Paris (Ed.) Natural lmiguage Generation and Computational I.inguistics, Dordrecht, Kluwer. Goerz, G. & Kesseler, M\" 1994. Anytime Algorithms for Speech Parsing? Proceedings of COUNG '94, Kyoto.", | |
| "links": null | |
| }, | |
| "BIBREF8": { | |
| "ref_id": "b8", | |
| "title": "Un algorithme d'analyse pour les granunaires d'arbres adjoints", | |
| "authors": [ | |
| { | |
| "first": "F", | |
| "middle": [], | |
| "last": "Issac", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1994, | |
| "venue": "Proceedings of the 1hird International TAG Workshop", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Issac, F., 1994. Un algorithme d'analyse pour les granunaires d'arbres adjoints. In: Proceedings of the 1hird International TAG Workshop, Paris. Technical report TALANA-RT-94-01, Universite Paris 7.", | |
| "links": null | |
| }, | |
| "BIBREF9": { | |
| "ref_id": "b9", | |
| "title": "Disambiguation of Super Parts of Speech (or Supertags): Almost Parsing", | |
| "authors": [ | |
| { | |
| "first": "A", | |
| "middle": [ | |
| "K" | |
| ], | |
| "last": "Joshi", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "B", | |
| "middle": [], | |
| "last": "Srinivas", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1994, | |
| "venue": "Proceedings of COUNG'94", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Joshi, A.K, & Srinivas, B., 1994. Disambiguation of Super Parts of Speech (or Supertags): Almost Parsing. Proceedings of COUNG'94, Kyoto.", | |
| "links": null | |
| }, | |
| "BIBREF10": { | |
| "ref_id": "b10", | |
| "title": "Un algorithme d'analyse LR(O) pour les Gramrnaires d' Arbres Adjoints Lexicalisees", | |
| "authors": [ | |
| { | |
| "first": "A", | |
| "middle": [], | |
| "last": "Kinyon", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1997, | |
| "venue": "Proceedings of TAIR'97", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Kinyon, A\" 1997. Un algorithme d'analyse LR(O) pour les Gramrnaires d' Arbres Adjoints Lexicalisees. Proceedings of TAIR'97.", | |
| "links": null | |
| }, | |
| "BIBREF11": { | |
| "ref_id": "b11", | |
| "title": "A Forrnalism and a Parser for Lexicalized Dependency Grammars", | |
| "authors": [ | |
| { | |
| "first": "A", | |
| "middle": [], | |
| "last": "Nasr", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1995, | |
| "venue": "Proceedings of the Fourth International Workshop on Parsing Technologies", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Nasr, A., 1995, A Forrnalism and a Parser for Lexicalized Dependency Grammars, Proceedings of the Fourth International Workshop on Parsing Technologies, Prague.", | |
| "links": null | |
| }, | |
| "BIBREF12": { | |
| "ref_id": "b12", | |
| "title": "Linear Indexed Automata and Tabulation of TAG Parsing", | |
| "authors": [ | |
| { | |
| "first": "M.-J", | |
| "middle": [], | |
| "last": "Nederhof", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1998, | |
| "venue": "Proceedings of the First Workshop on Tabulation in Parsing and Deduction", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Nederhof, M.-J., 1998. Linear Indexed Automata and Tabulation of TAG Parsing. Proceedings of the First Workshop on Tabulation in Parsing and Deduction, Paris.", | |
| "links": null | |
| }, | |
| "BIBREF13": { | |
| "ref_id": "b13", | |
| "title": "Multiset-Valued Linear Index Grammars: lmposing Dominance Conslraints on Derivations", | |
| "authors": [ | |
| { | |
| "first": "0", | |
| "middle": [], | |
| "last": "Rambow", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1994, | |
| "venue": "Proceedi11gs ofthe COUNG'94 Conference", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Rambow, 0., 1994. Multiset-Valued Linear Index Grammars: lmposing Dominance Conslraints on Derivations. Proceedi11gs ofthe COUNG'94 Conference, Kyoto.", | |
| "links": null | |
| }, | |
| "BIBREF14": { | |
| "ref_id": "b14", | |
| "title": "Parsing Slrategies with \"Lexicalized\" Grammars: Applications to Tree-Adjoining Grammars", | |
| "authors": [ | |
| { | |
| "first": "Y", | |
| "middle": [], | |
| "last": "Schabes", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "A", | |
| "middle": [], | |
| "last": "\" Abeille", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "A", | |
| "middle": [ | |
| "K" | |
| ], | |
| "last": "Joshi", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1988, | |
| "venue": "Proceedings of COUNG'88", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Schabes, Y\" Abeille, A. & Joshi, A.K., 1988. Parsing Slrategies with \"Lexicalized\" Grammars: Applications to Tree- Adjoining Grammars. Proceedings of COUNG'88, Budapest.", | |
| "links": null | |
| }, | |
| "BIBREF15": { | |
| "ref_id": "b15", | |
| "title": "An Earley-Type Parsing Algorithm for Tree Adjoining Grammars", | |
| "authors": [ | |
| { | |
| "first": "Y", | |
| "middle": [], | |
| "last": "Schabes", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "A", | |
| "middle": [], | |
| "last": "Joshi", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1988, | |
| "venue": "Proceedings of ACL'88", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Schabes, Y., & Joshi, A.K\" 1988. An Earley-Type Parsing Algorithm for Tree Adjoining Grammars. Proceedings of ACL'88, Buffalo.", | |
| "links": null | |
| }, | |
| "BIBREF16": { | |
| "ref_id": "b16", | |
| "title": "Two Recent Developments in Tree Adjoining Grammars: Semantics and Efficient processing", | |
| "authors": [ | |
| { | |
| "first": "Y", | |
| "middle": [], | |
| "last": "Schabes", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "A", | |
| "middle": [], | |
| "last": "Joshi", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1990, | |
| "venue": "Proceedings ofthe Third DARPA Speech andNatural Language Workshop", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Schabes, Y. & Joshi, A.K\" 1990. Two Recent Developments in Tree Adjoining Grammars: Semantics and Efficient processing. Proceedings ofthe Third DARPA Speech andNatural Language Workshop.", | |
| "links": null | |
| }, | |
| "BIBREF17": { | |
| "ref_id": "b17", | |
| "title": "Determ1nistic Left to Right Parsing of Tree Adjoining Languages", | |
| "authors": [ | |
| { | |
| "first": "Y", | |
| "middle": [], | |
| "last": "Schabes", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "K", | |
| "middle": [], | |
| "last": "Vijay-Shanker", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1990, | |
| "venue": "Proceedings of ACL'90", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Schabes, Y. & Vijay-Shanker, K, 1990. Determ1nistic Left to Right Parsing of Tree Adjoining Languages. Proceedings of ACL'90, Pittsburgh.", | |
| "links": null | |
| }, | |
| "BIBREF18": { | |
| "ref_id": "b18", | |
| "title": "Tree Insertion Grammar: A Cubic-Time Parsable Formalism That Lexicalizes Context-Free Grammar Without Changing the Trees Produced", | |
| "authors": [ | |
| { | |
| "first": "Y", | |
| "middle": [], | |
| "last": "Schabes", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "R", | |
| "middle": [], | |
| "last": "Waters", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1994, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Schabes, Y. & Waters, R.C\" 1994. Tree Insertion Grammar: A Cubic-Time Parsable Formalism That Lexicalizes Context-Free Grammar Without Changing the Trees Produced, Technical Report TR-94-13, Mitsubishi Eleclric Research Laboratories, Cambridge (MA).", | |
| "links": null | |
| }, | |
| "BIBREF19": { | |
| "ref_id": "b19", | |
| "title": "A Preferential Pattern-seeking Semantics for Natural Language lnference", | |
| "authors": [ | |
| { | |
| "first": "Y", | |
| "middle": [], | |
| "last": "Wilks", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1975, | |
| "venue": "Artificial Intelligence", | |
| "volume": "6", | |
| "issue": "", | |
| "pages": "53--74", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Wilks, Y\" 1975. A Preferential Pattern-seeking Semantics for Natural Language lnference. Artificial Intelligence, vol. 6, pp. 53-74.", | |
| "links": null | |
| } | |
| }, | |
| "ref_entries": { | |
| "FIGREF0": { | |
| "uris": null, | |
| "text": "Semantic propagation through substitution (''NO\" node of the PP group) enables selection of right furcation on \"N\", because of compatibility between :artifact and :omament features. [:act :openJ \u2022\u2022\u2022\u2022 \"\"\".\" .\u2022 \"\" ... \"\".\" \u2022\u2022 *V [telic = :open]", | |
| "type_str": "figure", | |
| "num": null | |
| } | |
| } | |
| } | |
| } |