UDPipe Chinese model is so bad at tokenization. I need to manaually seperate a doc into multiple paragraphs, then iteratively seperate each paragraph into multiple sentences, then iteratively tokenize each sentence with jiebaR. Then I need to feed the result into udpipe to go on tagging and parsing.
I read the official documentation and tried a lot, no luck. I'm not familiar with R.
Many thanks!
UDPipe Chinese model is so bad at tokenization. I need to manaually seperate a doc into multiple paragraphs, then iteratively seperate each paragraph into multiple sentences, then iteratively tokenize each sentence with jiebaR. Then I need to feed the result into udpipe to go on tagging and parsing.
I read the official documentation and tried a lot, no luck. I'm not familiar with R.
Many thanks!