Let me explain the reason a little bit. I am working in a Chinese bot and defined lots of concepts, which contains many more-than-five-words-long sequence.
As you might know, Chinese words are not naturally segmented. There are many tools try to segment Chinese words as precisely as possible for NLP purpose. I have been tried those tools with CS, and decided to abandon the approach, as I need 100% precise (or as I want which is of course impossible).
What I am doing is working on Character basis for Chinese. So it’s very easy to get to the 5-words limit, as many Chinese word (or phrase) contains 6-8 Character.
So is it possible to lift the max? I know it might affect performance. Or as an option that I can set when I build CS, that would be good enough for we to work on. So that we can define every phrase in CS, and CS can match whatever we intend.

RSS AI Zone

