WelCome
Akiyo Nadamoto's Home Page
Japanese
National Institute of Information and Communications Technology(NICT)
Knowledge Creating Communication Research Center
Computational Linguistics Group
Language Grid Rpoject
〒619-0289 3-5 Hikari-dai, Seika-cho, Soraku-Gun, Kyoto, Japan.
E-mail：nadamoto [at] nict.go.jp
Research
Publications
Research
2006～ Language Grid Project
The Language Grid offers two main benefits:
Ability to combine language resources (ex. bilingual dictionaries) or language processing functions (ex. machine translators).
Ability to add own language resources to create new language services for their own intercultural activities.
The Language Grid is an infrastructure that is built on the top of the Internet. It allows a better understanding of Internet contents written in different languages and by people from different countries. In addition, the Language Grid allows users to easily develop new language services by combining existing ones to satisfy their needs.
2002 - 2005 Bilingual Comparative Web Browser Based On Content-Synchronization and Viewpoint Retrieval
We propose a new way of browsing bilingual web sites through concurrent browsing with automatic similar-content synchronization and viewpoint retrieval facilities. Our prototype browser system is called the Bilingual Comparative Web Browser (B-CWB) and it concurrently presents bilingual web pages in a way that enables their contents to be automatically synchronized. The B-CWB allows users to browse multiple web news sites concurrently and compare their viewpoint of news articles written in different languages (English and Japanese). Our viewpoint retrieval is based on similar and different detection. We described categorizing pages in terms of viewpoint: the entire similarity, the content difference, and subject difference. Content synchronization means that user operation (scrolling or clicking) on one web page does not necessarily invoke the same operations on the other web page to preserve similarity of content between the multiple web pages. For example, scrolling a web page may invoke passage-level viewpoint retrieval on the other web page. Clicking a web page (and obtaining a new web page) invokes page-level viewpoint retrieval within the other site's pages through the use of an English-Japanese dictionary.
1998～2005 Transforming Web Content into TV-program-like Content Based on Automatic Transformation of Dialog
Web2Talkshow transforms declarative-based web content into humorous dialog-based TV-program-like content that is presented through a character agent and synthesized speech. Web2Talkshow enables users to get web content in a way similar to watching TV. The system transforms web content into dialog-based TV-program-like content based on the topic structure in the original web content. We formalized the TV-program-like content, the pre-processing part, the scenario part, and the direction part, to allow content to be reused. For the scenario part, we created methods for extracting the topic structure and for generating a humorous dialog based on the topic structure. With Web2Talkshow, users can get desired web content easily and in a user-friendly way while being able to continue working on other tasks; thus, it will be much like watching TV.