A Robust Parser Based on Syntactic Information

A Robust Parser Based on Syntactic Information Kong Joo Lee Cheol Jung Kweon Jungyun Seo Gil Chang Kim Department of Computer Scinence and CAIR Korea Advanced Institute of Science and Technology Taejon, Korea 305-701 {kjlee,cjkwn}@csone.kaist.ac.kr Abstract An extragrammatical sentence is what a normal parser fails to analyze. It is important to recover it using only syn- tactic information although results of recovery are better if semantic factors are considered. A general algorithm .for least-errors recognition, which is based only on syntactic information, was pro- posed by G. Lyon to deal with the ex- tragrammaticality. We extended this algorithm to recover extragrammatical sentence into grammatical one in run- ning text. Our robust parser with re- covery mechanism - extended general algorithm for least-errors recognition - can be easily scaled up and modified be- cause it utilize only syntactic informa- tion. To upgrade this robust parser we proposed heuristics through the analy- sis on the Penn treebank corpus. The experimental result shows 68% ,~ 77% accuracy in error recovery. 1 Introduction ExtragrammaticM sentences include patently un- grammatical constructions as well as utterances that may be grammaticMly acceptable but are be- yond the syntactic coverage of a parser, and any other difficult ones that are encountered in parsing (Carbonell and Hayes, 1983). I am sure this is what he means. This is, I am sure, what he means. The progress of machine does not stop even a day. Not even a day does the progress of machine stop. Above examples show that people are used to write same meaningful sentences differently. In addition, people are prone to mistakes in writing sentences. So, the bulk of written sentences are open to the extragrammaticality. In the Penn treebank tree-tagged cor- pus(Marcus, 1991), for instance, about 80 per- cents of the rules are concerned with peculiar sentences which include inversive, elliptic, paren- thetic, or emphatic phrases. For example, we can drive a rule VP ---, vb NP comma rb comma PP from the following sentence. ( (S The same jealousy can breed confusion, however, in the absence of any authoriza- tion bill this year. (NP The/dr (ADJP ea~e/jj) jealousy/nn) can/md (VP breed/vb (NP confusion/nn) ,/, however/rb ,/, (PP in/in (NP (NP the/dr absence/nn) (PP orlon (NP anyldt auZhorization/nnbi11/nn)) (NP this/dt yearlnn))))) ./.) A robust parser is one that can analyze these extragrammaticalsentences without failure. How- ever, if we try to preserve robustness by adding such rules whenever we encounter an extra- grammatical sentence, the rulebase will grow up rapidly, and thus processing and maintaining the excessive number of rules will become inefficient and impractical. Therefore, extragrammatical sentences should be handled by some recovery mechanism(s) rather than by a set of additional rules. Many researchers have attempted several tech- niques to deal with extragrammatical sentences such as Augmented Transition Network(ATN) (Kwasny and Sondheimer, 1981), network-based semantic grammar (Hendrix, 1977), partial pat- tern matching (Hayes and Mouradian, 1981), con- ceptual case frame (Schank et al., 1980), and mul- tiple cooperating methods (Hayes and Carbonell, 1981). Above mentioned techniques …

Content truncated to 500 words