Stanford InfoLab Publication Server

Merging Hierarchies Using Object Placement

Zhao, Kai and Ikeda, Robert and Garcia-Molina, Hector (2007) Merging Hierarchies Using Object Placement. Technical Report. Stanford.




The main challenge in integrating two hierarchies is determining the correspondence between the nodes and edges of each hierarchy. Traditionally, the correspondence is determined by comparing the text associated with each edge. In this paper we instead use the placement of objects present in both hierarchies to infer how the hierarchies relate. We present an algorithm that, given a hierarchy with known facets (label-value pairs that define what objects are placed under an edge), determines feasible facets for a second hierarchy, based on shared objects. We present a second algorithm that uses the discovered facets to copy objects that only appear in the second hierarchy into the first one, to arrive at a more comprehensive hierarchy. We present experimental results that illustrate the usefulness of our approach, and how noise in our data can be handled.

Item Type:Techreport (Technical Report)
Subjects:Computer Science > Data Integration and Mediation
Related URLs:Project Homepage
ID Code:798
Deposited By:Import Account
Deposited On:01 Mar 2007 16:00
Last Modified:10 Dec 2008 18:00

Download statistics

Repository Staff Only: item control page