File size: 2,665 Bytes
dc78b20 |
1 |
we are concerned with surface-syntactic parsing of running text. our main goal is to describe syntactic analyses of sentences using dependency links that show the head-modifier relations between words. in addition, these links have labels that refer to the syntactic function of the modifying word. a simplified example is in figure 1, where the link between i and see denotes that i is the modifier of see and its syntactic function is that of subject. similarly, a modifies bird, and it is a determiner. first, in this paper, we explain some central concepts of the constraint grammar framework from which many of the ideas are derived. then, we give some linguistic background to the notations we are using, with a brief comparison to other current dependency formalisms and systems. new formalism is described briefly, and it is utilised in a small toy grammar to illustrate how the formalism works. finally, the real parsing system, with a grammar of some 2 500 rules, is evaluated. the parser corresponds to over three man-years of work, which does not include the lexical analyser and the morphological disambiguator, both parts of the existing english constraint grammar parser (karlsson et al., 1995). the parsers can be tested via www'.voutilainen and juha heikkild created the original engcg lexicon. we are using atro voutilainen's (1995) improved part-of-speech disambiguation grammar which runs in the cg-2 parser. the parsers can be tested via www'. we are concerned with surface-syntactic parsing of running text. in this paper, we have presented some main features of our new framework for dependency syntax. however, the comparison to other current systems suggests that our dependency parser is very promising both theoretically and practically. our work is partly based on the work done with the constraint grammar framework that was originally proposed by fred karlsson (1990). for instance, the results are not strictly comparable because the syntactic description is somewhat different. the evaluation was done using small excerpts of data, not used in the development of the system. our main goal is to describe syntactic analyses of sentences using dependency links that show the head-modifier relations between words. the distinction between the complements and the adjuncts is vague in the implementation; neither the complements nor the adjuncts are obligatory. means that a nominal head (nom-head is a set that contains part-of-speech tags that may represent a nominal head) may not appear anywhere to the left (not *-1). this "anywhere" to the left or right may be restricted by barriers, which restrict the area of the test. |