Schema Evolution publication categorizer

Ontology matching is a crucial task of enabling interoperation between Web applications using different but related ontologies. Due to the size and the monolithic nature, large-scale ontologies regarding real world domains cause a new challenge to current ontology matching techniques. In this paper, we propose a method for partition-based block matching that is practically applicable to large class hierarchies, which are one of the most common kinds of large-scale ontologies.

Internet marketplaces are now faced with new
challenges that arise from the need to seamlessly
integrate enormous number of product catalogs from
different sources. In order to help users find products
efficiently, Internet shops provide hierarchies of product
catalogs (called category hierarchies). However, the
absence of robust models (and well understood
semantics) for product catalogs and their hierarchies
severely impairs our ability to systematically support
structured operations.
In this paper, we present an extended catalog model

One of the main challenges in integrating two hierarchies (e.g., of books or web pages) is determining the correspondence between the edges of each hierarchy. Traditionally, this process, which we call hierarchy matching, is done 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 address the problem of integrating objects from a source taxonomy into a master taxonomy. This problem is not only currently pervasive on the web, but also important to the emerging semantic web. A straightforward approach to automating this process would be to learn a classifier that can classify objects from the source taxonomy into categories of the master taxonomy. The key insight is that the availability of the source taxonomy data could be helpful to build better classifiers for the master taxonomy if their categorizations have some semantic overlap.

Many approaches have been proposed for the
of mapping categories (classes) from a source
to classes in a master taxonomy. Most of techniques,
however, ignore the hierarchical structure
taxonomies. In this paper, we propose a maximum likelihood
based framework which exploits the hierarchical
structure to obtain a more natural mapping between
source classes and the master taxonomy. Furthermore,
unlike previous work, our technique also inserts
classes into appropriate places of the master
creating new categories if required. We evaluate approach
on text and hyperspectral datasets.

We address the problem of unsupervised matching of schema information from a large number of data sources into the