Concept lattice compression based on K-means

Formal concept analysis has been applied as a tool for knowledge expression and acquisition. However, the huge concept lattice makes the hidden knowledge difficult to understand. This paper proposes a method to compress a concept lattice using K-means clustering. Firstly, the similarity measure between formal concepts is obtained through the importance degree of each attribute and object, and then, the concepts are clustered by K-means clustering. Finally, we define a K-deletion transformation to realize the compression of concept lattice.