Open yangovoyang opened 6 years ago
While I haven't use SEMPRE with languages outside English, It should work with Chinese if the language analyzer is set properly. SEMPRE uses Stanford CoreNLP to analyze texts.
CoreNLP supports Chinese, but the correct models have to be loaded and the correct parameters have to be set. Get the models from:
https://stanfordnlp.github.io/CoreNLP/human-languages.html
and modify the initModels
function of the file
src/edu/stanford/nlp/sempre/corenlp/CoreNLPAnalyzer.java
to point to the Chinese models. Other parts of the code would probably need to be changed as well (e.g., the Chinese model does not have lemmatization)
I am a newbee in this research field, and before I dive into this toolkit, could anyone tell me whether it supports Chinese? So that I can write my rules to parse Chinese uttrance into the logic presentations. My sincere thanks!