Automating Creation of Hierarchical Faceted Metadata Structures

We describe Castanet, an algorithm for automatically generating hierarchical faceted metadata from textual descriptions of items, to be incorporated into browsing and navigation interfaces for large information collections. From an existing lexical database (such asWordNet), Castanet carves out a structure that reflects the contents of the target information collection; moderate manual modifications improve the outcome. The algorithm is simple yet effective: a study conducted with 34 information architects finds that Castanet achieves higher quality results than other automated category creation algorithms, and 85% of the study participants said they would like to use the system for their work.