infiniflow
[Question]: use of graphRAG
Describe your problem
The default key of graphRAG is "person"、“location”、"time"，however,my content of pdf file is "Chapter 1 Introduction to Computers"、"Chapter 2 What Linux is and how to learn it"...haw should i set the keywords of grapRAG. Meanwhile,how is the number of tokens(8192 default) should be set.
I don't think it's suitable for graphRAG.
What about using General to parse?
I don't think it's suitable for graphRAG. What about using
Generalto parse?
OK，the General to parse is fine. However, i just want to know that how can i extract the keywords of graphRAG in the file,except the "person"、“location”、"time"...
person, location..., these are entity types. You could define the entity types you want just by editing.
person, location..., these are entity types. You could define the entity types you want just by editing.
hey Kevin, sorry for disturbing. i wanna know that if i have parsed files and then i redefined the entity types, should i parse the files again? cause i found there's no difference in Knowledge graph after that.
