I want to build a tool that generates sensible syllogisms. An example of a syllogisms is: all A are B. all C are A. all C are B). I want the triplet (A, B, C) to be semantically related to each other in the way described by the syllogisms. That is, I would like my tool to generate 'a=humans; b=mortal; c=greeks' and not 'a=chickens; b=burgers; c=frogs'.
There's a syllogism generator online (http://krypton.mnsu.edu/~jp5985fj/courses/609/Logic/Silly%20Syllogisms.htm) but it doesn't generate syllogisms that are semantically related, it generates random terms for A, B, and C which may or may not be related.
My question is, in NLP, are there any research papers for generating semantically valid syllogisms? What topics would I need to research to build this tool?