A domain-independent, ontology-agnostic approach to leveraging unstructured data sources

ABSTRACT

The vast accumulation of unstructured data within various types of documents in recent years has escalated the need for more efficient and effective management of available data. An effective tool for unstructured data management can vastly benefit business enterprises as well as support the understanding of volumes of scientific data in academic and scholarly works. This paper presents a simple tool that would make such data more actionable by allowing for an automated mechanism to support sifting through vast amounts of data and understand simplistic correlations between the data sources. The system is designed to extract meaning from data sources in a user-directed fully automated approach. Flexibility and robustness was incorporated by allowing for the accommodation of multiple data source types and leveraging the WordNet ontology for subsequent data analysis.

REFERENCES

• 1: G. Weglarz, "Two Worlds of Data -- Unstructured and Structured," {Online Document}, {2007 Sep 15}, Available at http://www.dmreview.com/article_sub.cfm?articleId=100916 1.
• 2: A. Adshead, "Unearth The Benefits of Buried Data," {Online Document}, June 2007, {2007 Sep 10}, Available at http://www.computerweekly.com/Articles/2007/06/15/224787/unearth-the-benefits-of-buried-data.htm.
• 3: R. Rao, "From Unstructured Data to Actionable Intelligence," {Online Document}, Nov 2003 {2007 Sep 18}, Available at http://www.ramanarao.com/papers/rao-itpro-2003-11.pdf.
• 4: Martin Hepp, Possible Ontologies: How Reality Constrains the Development of Relevant Ontologies, IEEE Internet Computing, v.11 n.1, p.90-96, January 2007 [doi>10.1109/MIC.2007.20]
• 5: A. Maedche and S. Staab. 2000a. Discovering conceptual relations from text. In Proceedings of ECAI-2000. IOS Press, Amsterdam.
• 6: N. Roy and M. Musen, "PROMPT: Algorithm and Tool for Automated Ontology Merging and Alignment," {Online Document}, {2007 Oct 16}, Available at http://smi.stanford.edu/smi-web/reports/SMI-2000-0831.pdf.
• 7: Peter Buneman , Susan Davidson , Gerd Hillebrand , Dan Suciu, A query language and optimization techniques for unstructured data, Proceedings of the 1996 ACM SIGMOD international conference on Management of data, p.505-516, June 04-06, 1996, Montreal, Quebec, Canada
• 8: Robert M. Losee, Browsing mixed structured and unstructured data, Information Processing and Management: an International Journal, v.42 n.2, p.440-452, March 2006 [doi>10.1016/j.ipm.2005.02.001]
• 9: J. Mazon and J. Tullion, "Using WordNet Ontology to Automatically Enrich Dimension Hierarchies in a Data Warehouse," {Online Document}, {2007 Oct 4}, Available at http://www.i3s.unice.fr/odbis2005/Using%20WordNet%20Ontology%20.pdf
• 10: Michael Collins , Charles Reynolds , Christine Le , Cihan Varol , Coskun Bayrak, Automated data verification in a format-free environment, ACM SIGSOFT Software Engineering Notes, v.31 n.2, March 2006 [doi>10.1145/1118537.1118539]

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.0 General

Additional Classification:
  E. Data
  E.5 FILES
      Subjects: Organization/structure

General Terms:
Algorithms, Design, Standardization

Keywords:
data analysis, entity extraction, information extraction, ontology, unstructured data