This paper introduces a tool \Bonsai which supports human in annotating corpora with morphosyntactic information, and in retrieving syntactic structures stored in the database. Integrating annotation and retrieval enables users to annotate a new instance while looking back at the already annotated sentences which share the similar morphosyntactic structure. We focus on the retrieval part of the system, and describe a method to decompose a large input query into smaller ones in order to gain retrieval efficiency. The proposed method is evaluated with the Penn Treebank corpus, showing significant improvements.

In Japanese constructions of the form [N1 no Adj N2], the adjective Adj modifies either N1 or N2. Determing the semantic dependencies of adjective in such phrase is an important task for machine translation. This paper describes a method for determining the adjective dependency in such constructions using decision lists, and inducing decision lists from training contexts with correct semantic dependencies and without. Based on evaluation, our method is able to determine adjective dependency with an precision of about 94%. We further analyze rules in the induced decision lists and examine effective features to determine the semantic dependencies of adjectives.

Bracketed corpora are a very useful resource for natural language processing, but hard to build efficiently, leading to quantitative insufficiency for practical use. Disparities in morphological information, such as word segmentation and part-of-speech tag sets, are also troublesome. An application specific to a particular corpus often cannot be applied to another corpus. In this paper, we sketch out a method to build a corpus that has a fixed syntactic structure but varying morphological annotation based on the different tag set schemes utilized. Our system uses a two layered grammar, one layer of which is made up of replaceable tag-set-dependent rules while the other has no such tag set dependency. The input sentences of our system are bracketed corresponding to structural information of corpus. The parser can work using any tag set and grammar, and using the same input bracketing, we obtain corpus that shares partial syntactic structure.

This paper presents a new formalization of probabilistic GLR language modeling for statistical parsing. Our model inherits its essential features from Briscoe and Carroll’s generalized probabilistic LR model, which obtains context-sensitivity by assigning a probability to each LR parsing action according to its left and right context. Briscoe and Carroll’s model, however, has a drawback in that it is not formalized in any probabilistically well-founded way, which may degrade its parsing performance. Our formulation overcomes this drawback with a few significant refinements, while maintaining all the advantages of Briscoe and Carroll’s modeling.

There is a big shift in MT R&D in this region after many large-scale projects conducted in the past ten years. Multi-lingual Machine Translation (MMT) project is one of the significant R&D projects that increased a great number of NLP related researchers and research activities which can be seen in the increasing number of the research institutes in the recent years. We learned a lot from the collaboration research across languages and we still hope that it will be a rigorous step for the future MT R&D in this region. Though the MT systems are still far from the extreme goal of the perfect translation, it can be observed that the MT systems are actually used to support information retrieval from the Internet.

A generalized LR parsing algorithm, which has been developed by Tomita [Tomita 86], can treat a context free grammar. His algorithm makes use of breadth first strategy when a conflict occcurs in a LR parsing table. It is well known that the breadth first strategy is suitable for parallel processing. This paper presents an algorithm of a parallel parsing system (PLR) based on a generalized LR parsing. PLR is implemented in GHC [Ueda 85] that is a concurrent logic programming language developed by Japanese 5th generation computer project. The feature of PLR is as follows: Each entry of a LR parsing table is regarded as a process which handles shift and reduce operations. If a process discovers a conflict in a LR parsing table, it activates subprocesses which conduct shift and reduce operations. These subprocesses run in parallel and simulate breadth first strategy. There is no need to make some subprocesses synchronize during parsing. Stack information is sent to each subprocesses from their parent process. A simple experiment for parsing a sentence revealed the fact that PLR runs faster than PAX [Matsumoto 87][Matsumoto 89] that has been known as the best parallel parser.