SHACHI - Language Resource Metadata Database

A full-text news paper article database containing data from both national and local editions of Yomiuri Newspaper articles published in 2007. The data is provided in text format. CSV-format files are also available from Nihon Database Kaihatsu Co., Ltd.

A full-text news paper article database containing data from both national and local editions of Yomiuri Newspaper articles published in 2006. The data is provided in text format. CSV-format files are also available from Nihon Database Kaihatsu Co., Ltd.

The database contains about 9,500 newspaper articles from the Daily Yomiuri (written in English) published in 2007. The data is exclusively for research and academic use and is intended to support development and studies in such fields as linguistics, informatics or media study. All the data is provided in CSV format.

The database contains about 9,000 newspaper articles from the Daily Yomiuri (written in English) published in 2006. The data is exclusively for research and academic use and is intended to support development and studies in such fields as linguistics, informatics or media study. The data is provided in CSV format.

A full-text news paper article database containing data from The Daily Yomiuri newspaper articles published in 2007. This is the only English newspaper article database published in Japan. The data is provided in text format. CSV-format files are also available from Nihon Database Kaihatsu Co., Ltd.

A full-text news paper article database containing data from The Daily Yomiuri newspaper articles published in 2006. This is the only English newspaper article database published in Japan. The data is provided in text format. CSV-format files are also available from Nihon Database Kaihatsu Co., Ltd.

MUST1 is a database of Japanese compound functional expressions and their example usage. It is designed to support studies on computational processing of Japanese compound expressions. It contains 337 entries (125 compound functional expressions selected based on "Gendaigo Hukugouji Youreishu" by National Institute for Japanese Language and their variants). Each entry has at most 50 example sentences. Note that the package "MUST1-dist" does not include the text data from Mainichi Shimbun newspaper articles 95 CD-ROM.

The UAM Spanish Treebank is a corpus of syntactically annotated Spanish sentences extracted from Spanish newspapers. The current version contains 1600 sentences and their goal is to have 5,000 sentences annotated. The sentences were annotated for syntactic categories (i.e. POS), syntactic functions, syntactic features (e.g. number, gender, tense. etc.) and semantic features (e.g. HUMAN, TIME, etc.). The annotation format is a vertical and indented format close to the PROTEUS format (http://nlp.cs.nyu.edu/index.shtml), and the Penn Treebank schema was used for annotating null elements.

OpenMWE is a set of language resources of multiword expressions (MWEs) and it is available as open source software. It is aimed to provide language resources for studying and developing technologies for computational processing of multiword expressions. The OpenMWE corpus, one of the OpenMWE resources, is a set of Japanese example sentences containing MWEs, focusing on MWEs with ambiguous meanings. Each entry MWE is given more or less 1000 example sentences.

The list contains about 3600 Japanse idioms appearing in 5 different Japanese sources (2 Japanese dictionaries for elementary school children, two standard Japanese idiom dictionaries and a Japanese idiom dictionary for children). The comparative list tells users which idioms appear in what resource(s).