The Formal Concept Analysis of the Documents Clusters
Hui-Feng Shi, Qiang Hua, Po Zhang · 2007
In this paper, an method based on formal concept analysis for explaining the document clustering result is presented. In order to decrease the number of objects in formal concept analysis, the documents is clustered by high dimensionala sparse clustering approach. The clusters are considered as the objects of formal context; The attributes set is the characters set of centroid vectors. The concept lattice of this formal context is builded by the Ganter algorithm. From the line graph of the concept lattice with respect to the formal context, we can explain that why some documents are clustered, and find the content of those documents in the each cluster.