Treebank Semantics processes constituency tree annotations and returns logic based meaning representations.
There is documentation that describes the approach and implementation in detail.
The system is distributed freely.
A web based interface is provided to the Treebank Semantics Parsed Corpus (TSPC), which is a testing ground for the system: Meaning representations are built automatically by the system from ‘live’ treebank files.
There is also an online demo that you are welcome to try.
Development is funded by the Japan Society for the Promotion of Science (JSPS) (KAKENHI-PROJECT-19K00541), and the National Institute for Japanese Language and Linguistics (Development of and Linguistic Research with a Parsed Corpus of Japanese Project).
Previously, project funding came from the Japan Science and Technology Agency (JST) (Synthesis of Knowledge for Information Oriented Society), and the Japan Society for the Promotion of Science (JSPS) (KAKENHI-PROJECT-15K02469).
Email: ajb129 __AT__ hotmail __DOT__ com.
Last updated: Nov 27, 2020