Deep text analytics for dark data in Oil and Gas
As a recent study by Gartner (Roundup of Energy and Utilities Research, 2013) highlights oil and gas companies are increasingly focusing on analytics to move from reactive reporting to “more advanced business processes” that can solve problems before they occur. Many of these initiatives are leveraging the “dark data” that reside in multiple systems and across technology domains.
Deep semantic technology is being used by some of the world’s largest oil and gas companies to leverage this information. This “dark data” is commonly present in any knowledge base. In the oil and gas sector, it could be drilling or exploratory information from a site collected by third party contractors, or information in collaboration platforms that lies in an information back log untagged and therefore inaccessible by search or other information discovery applications.
By definition, oil and gas operations generate massive volumes of data, from seismic data, to detailed drilling reports and well data, as well as other multi-disciplinary information on geology, hydraulics, etc. Competitive information, including social media posts, news reports are also a valuable source of information for understanding environmental conditions on project sites, as well as emerging political information that could impact worker safety and supply chains.
The use of semantic technology provides analysts who must pore over this data with a number of advantages in the process of mining large volumes of data for useful intelligence:
• Define and categorize content in the context of the organization managing it.
• Supply reasoning and contextual knowledge (this means X instead of Y) instead of leaving it undetermined and therefore not accessed by search or other discovery methods.
• Leverage natural language processing to take advantage of social media information where slang and idiomatic expressions are commonly used.
• Connect similar concepts in disparate sources and complex knowledge bases to make information comprehensible and useable by all applications.
• Discover patterns in information and capture weak signals that enhance predictive analysis models
If you plan to be in Houston for the upcoming International Conference on Petroleum Data, Integration & Data Management conference, we hope you’ll join us in booth 411 for a live demo of our semantic software.