Takako Aikawa

A U-shaped Relative Clause Attachment Preference in Japanese

This paper presents results from a self-paced reading experiment in Japanese investigating attachment preferences for relative clauses to three ensuing potential nominal heads. Similar to previous results from the processing of English, Spanish and German, we observed the following non-monotonic preference ordering among the three attachment sites: most local, least local, intermediate. We discuss the result in light of two types of parsing models: models that only consider attaching a modifier to candidate sites whose lexical heads have already been encountered, and models in which predicted categories are also considered as possible modification sites. We contend that the preference to attach to the least local site over the intermediate site argues against the first type of model, and supports the second type of model with a factor such as predicate proximity or anaphor resolution driving the preference to attach the RC to the least local candidate site.

The Impact of Crowdsourcing Post-editing with the Collaborative Translation Framework

This paper presents a preliminary report on the impact of crowdsourcing post-editing through the so-called ”Collaborative Translation Framework” (CTF) developed by the Machine Translation team at Microsoft Research. We first provide a high-level overview of CTF and explain the basic functionalities available from CTF. Next, we provide the motivation and design of our crowdsourcing post-editing project using CTF. Last, we present the results from the project and our observations. Crowdsourcing translation is an increasingly popular-trend in the MT community, and we hope that our paper can shed new light on the research into crowdsourcing translation.

English-Japanese example-based machine translation using abstract linguistic representations

This presentation describes an example-based English-Japanese machine translation system in which an abstract linguistic representation layer is used to extract and store bilingual translation knowledge, transfer patterns between languages, and generate output strings. Abstraction permits structural neutralizations that facilitate learning of translation examples across languages with radically different surface structure characteristics, and allows MT development to proceed within a largely language-independent NLP architecture. Comparative evaluation indicates that after training in a domain the English-Japanese system is statistically indistinguishable from a non-customized commercially available MT system in the same domain.

Multilingual sentence generation

This paper presents an overview of a robust, broad-coverage, and application-independent natural language generation system. It demonstrates how the different language generation components function within a multilingual Machine Translation (MT) system, using the languages that we are currently working on (English, Spanish, Japanese, and Chinese). Section 1 provides a system description. Section 2 focuses on the generation components and their core set of rules. Section 3 describes an additional layer of generation rules included to address application-specific issues. Section 4 provides a brief description of the evaluation method and results for the MT system of which our generation components are a part.

Two Types of Zi-Verbs in Japanese

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. American Association of Teachers of Japanese is collaborating with JSTOR to digitize, preserve and extend access to The Journal of the Association of Teachers of Japanese.