Functional Structures for Parsing Dependency Constraints
نویسندگان
چکیده
In the theory of computation a new viewpoint is digging in: to compute is to pin down the constraints that hold in a given problem domain, and a goal for computation. It is up to an interpreter to perform search for the goal in the problem domain. The result of computation follows then indirectly from the search process. Strongly pronounced and in wide use this vantage point becomes in Prolog. Recently fresh views of parsing as constraint systems have also surfaced, such as FUG (Kay, 1985; Karttunen and Kay, 1985), LFG (Bresnan, 1978), and PATR-II (Schieber, 1985). In these languages, a user writes only grammatical constraints and need not import control instructions. The interpreter searches for a grammatical configuration that "explains" a given input sentence without violating the constraints. These new grammars advocate yet another, more abstract departure from procedural description. Grammars for parsing have predominantly used generative rewrite rules. The ideological underpinning of parsing has been in the past that of the emulation of generative tllstories of configurations. The new formalisms express grammars as functional structures. We have defined a language DPL (Dependency Parsing Language) to meet the needs of parsing a highly inflectional, agglutinating language (Nelimarkka et al., 1984). The language enforces dependency approach which accords better than phrase structure rules with the needs of non-configurational languages. DPL language and its compiler constitutes .just one component of a language-based environment we have implemented for the development of parsers (Lehtota et al., 1985). In DPL, a grammar is comprised of functions, relations, and automata. The automata, which control the parsing process, have compelled a person who writes grammar to heed control unwanted extent. This paper describes a high-level language FUNDPL (FUNctional DPL) we have designed on top of DPL. In FUNDPL, a grammar is built out of functions, relations, and functional structures. FUNDPL is a constraint system which liberates a grammar writer from control anxieties.
منابع مشابه
An improved joint model: POS tagging and dependency parsing
Dependency parsing is a way of syntactic parsing and a natural language that automatically analyzes the dependency structure of sentences, and the input for each sentence creates a dependency graph. Part-Of-Speech (POS) tagging is a prerequisite for dependency parsing. Generally, dependency parsers do the POS tagging task along with dependency parsing in a pipeline mode. Unfortunately, in pipel...
متن کاملStochastic Dependency Parsing of Spontaneous Japanese Spoken Language
This paper describes the characteristic features of dependency structures of Japanese spoken language by investigating a spoken dialogue corpus, and proposes a stochastic approach to dependency parsing. The method can robustly cope with inversion phenomena and bunsetsus which don’t have the head bunsetsu by relaxing the syntactic dependency constraints. The method acquires in advance the probab...
متن کاملHybrid Constraints for Robust Parsing: First Experiments and Evaluation
In this paper we present IDEAL+, a parsing architecture for Italian, which pursues the goal of pairing robustness with deep linguistic analysis by extending a shallow processing kernel with a pool of hybrid constraints for the incremental identification of grammatical relations. The parsing output takes the form of dependency structures representing the full range of instantiated functional rel...
متن کاملLFG without C-structures
We explore the use of two dependency parsers, Malt and MST, in a Lexical Functional Grammar parsing pipeline. We compare this to the traditional LFG parsing pipeline which uses constituency parsers. We train the dependency parsers not on classical LFG f-structures but rather on modified dependency-tree versions of these in which all words in the input sentence are represented and multiple heads...
متن کاملتأثیر ساختواژهها در تجزیه وابستگی زبان فارسی
Data-driven systems can be adapted to different languages and domains easily. Using this trend in dependency parsing was lead to introduce data-driven approaches. Existence of appreciate corpora that contain sentences and theirs associated dependency trees are the only pre-requirement in data-driven approaches. Despite obtaining high accurate results for dependency parsing task in English langu...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1986