GIS: The Key to Unlocking the Unstructured Data Prize

Eric Hatleberg, Charles Basden, Brett Jones, Alejandro Mateos, and Philippe Steinthal
Schlumberger Information Solutions, Houston, TX

We typically regard data as structured (kept in a DBMS or standardized file format) or unstructured (anything you can imagine).

Structured data is highly constrained and controlled. Examples include logs, markers and seismic data. It is the traditional focus of information management solutions. There are many tools to create, maintain, search and analyze structured data however the flow of information in this channel is very wide but very shallow.

Conversely, unstructured data is loosely constrained and controlled. It includes reports, articles, and photos. This information is often directly tied to the decision making process and is growing at such a phenomenal rate that it threatens to overwhelm our IT systems. Our tools for unstructured data cluster around document creation and are weak on maintenance, search and analysis. This information flows deep, wide and strong but is often overlooked.

The dilemma for unstructured data is there are few basic organizing principles. A strategy that uses GIS technology to capture spatial information from unstructured data provides the breakthrough necessary to search and analyze our richest set of information. Additionally unstructured data is generally held in many dispersed locations across an organization. Therefore data management approaches must take a broader view. To be effective there must be a way to connect and evaluate multiple data sources.

An update on progress toward spatially enabling unstructured data will be demonstrated. Issues including integration with structured data, incorporation of spatial analysis capabilities and the use of various query and visualization tools will be discussed.