Michiko Kosaka

Covering Treebanks with GLARF

This paper introduces GLARF, a framework for predicate argument structure. We report on converting the Penn Treebank II into GLARF by automatic methods that achieved about 90% precision/recall on test sentences from the Penn Treebank. Plans for a corpus of hand-corrected output, extensions of GLARF to Japanese and applications for MT are also discussed.

A Multilingual Procedure for Dictionary-Based Sentence Alignment

This paper describes a sentence alignment technique based on a machine readable dictionary. Alignment takes place in a single pass through the text, based on the scores of matches between pairs of source and target sentences. Pairings consisting of sets of matches are evaluated using a version of the Gale-Shapely solution to the stable marriage problem. An algorithm is described which can handle N-to-1 (or 1-to-N) matches, for n ≥ 0, i.e., deletions, 1-to-1 (including scrambling), and 1-to-many matches. A simple frequency based method for acquiring supplemental dictionary entries is also discussed. We achieve high quality alignments using available bilingual dictionaries, both for closely related language pairs (Spanish/English) and more distantly related pairs (Japanese/English).

Chart-based transfer rule application in Machine Translation

Transfer-based Machine Translation systems require a procedure for choosing the set of transfer rules for generating a target language translation from a given source language sentence. In an MT system with many competing transfer rules, choosing the best set of transfer rules for translation may involve the evaluation of an explosive number of competing sets. We propose a solution to this problem based on current best-first chart parsing algorithms.

Automatic Recognition of Logical Relations for English, Chinese and Japanese in the GLARF Framework

We present GLARF, a framework for representing three linguistic levels and systems for generating this representation. We focus on a logical level, like LFGs F-structure, but compatible with Penn Treebanks. While less finegrained than typical semantic role labeling approaches, our logical structure has several advantages: (1) it includes all words in all sentences, regardless of part of speech or semantic domain; and (2) it is easier to produce accurately. Our systems achieve 90% for English/Japanese News and 74.5% for Chinese News -- these F-scores are nearly the same as those achieved for treebank-based parsing.

Transducing Logical Relations from Automatic and Manual GLARF

GLARF relations are generated from treebank and parses for English, Chinese and Japanese. Our evaluation of system output for these input types requires consideration of multiple correct answers.