Granular Computing and Modeling the Human Thoughts in Web Documents

摘要

The totality of human thoughts in a document set is modeled by a polyhedron. A point represents a THOUGHT, a simplex a CONCEPT, a connected component a COMPLETE CONCEPT, the simplicial structure the whole IDEA. The building block is the simplex; it represents the concept that is carried by a set of high frequency and nearby co-occurring keywords. The simplicial structure of the keywords provides an ”informal” formal language about human thoughts in a document set. The model theory of this language gives the desirable model.