Mining semantics for large scale integration on the web
Authors:
Open Access
- 1 December 2004
- journal article
- Published by Association for Computing Machinery (ACM) in ACM SIGKDD Explorations Newsletter
- Vol. 6 (2) , 67-76
Abstract
The Web has been rapidly "deepened" -- with myriad searchable databases online, where data are hidden behind query interfaces. Toward large scale integration over this "deep Web," we are facing a new challenge- With its dynamic and ad-hoc nature, such large scale integration mandates dynamic semantics discovery. That is, we must on-the-fly cope with "semantics" of dynamically discovered sources without pre-configured source-specific knowledge. To tackle this challenge, our initial works hinge on the insight that the large scale is itself also a unique opportunity: We observe that the desired "semantics" often connects to surface presentation characteristics, through some hidden regularities over many sources. Such regularities can be essentially leveraged in enabling semantics discovery. In particular, we report our evidences in three initial tasks for integrating the deep Web: interface extraction, schema matching , and query translation. Generalizing these specific evidences, we thus propose our "unified insight" of "mining" semantics for large scale integration by exploiting hidden regularities across holistic sources. Further, to fulfill the promise of such holistic mining, we discuss challenges toward its realization for dynamic semantics discovery. As our initial works as well as several related efforts have witnessed, we believe our unified insight, holistic mining for semantics discovery, is a promising methodology toward enabling large scale integration.Keywords
This publication has 13 references indexed in Scilit:
- Corpus-Based Schema Matching2005
- Structured databases on the webACM SIGMOD Record, 2004
- Discovering complex matchings across web query interfaces2004
- Understanding Web query interfaces2004
- An interactive clustering-based approach to integrating source query interfaces on the deep Web2004
- Constraint multiset grammars2002
- A survey of approaches to automatic schema matchingThe VLDB Journal, 2001
- Reconciling schemas of disparate data sources2001
- The Clio projectACM SIGMOD Record, 2001
- Database techniques for the World-Wide WebACM SIGMOD Record, 1998