Multi-national Oil and Gas Company

Case Study

  Industry Case Studies    All Case Studies

Saving Millions Using Content Analytics

“Traditional search assumes that end users know what they are looking for, or must enter the ‘right’ combination of words, in the ‘right’ order, to get the ‘right’ result. This factor is an issue in successfully creating a data set for text analytics.”

Concept Searching
Customer Location:
United Kingdom
Industry:
Energy and Utilities
Issue:

Inability to aggregate relevant and accurate content from diverse repositories, to improve decision making and reduce costs.

Products:

“The objective was to analyze and extract relevant information to improve geological decision making in the selection or abandonment of drilling sites. Access to the project included 5,000 geophysicists globally and required the integration of 52 taxonomies. The end result of the analysis represented millions of dollars in savings to the company.”

A highly scalable solution, easily usable by subject-matter experts, that can extract highly granular content in context from diverse repositories, to improve decision making by providing the data needed to solve specific business issues.

Benefits

  • Easily used by subject-matter experts, intuitive, and requires little training
  • Ability to extract and aggregate content from multiple taxonomies and repositories
  • Finds unknown content, making it available for analysis
  • Identifies inter-related and intra-related content automatically
  • Creates a clean data set, by eliminating noise and content of no value
  • Can be used by any business or artificial intelligence applications for manipulation and analysis

One of the leading global integrated oil and gas companies, this client provides fuel for transportation, energy for heat and light, lubricants to keep engines moving, and the petrochemicals products used to make everyday items as diverse as paints, clothes, and packaging.

This organization required more than traditional search. Traditional search assumes that end users know what they are looking for, or must enter the ‘right’ combination of words, in the ‘right’ order, to get the ‘right’ result. This factor is an issue in successfully creating a data set for text analytics. There are over 171,000 words in the English language. Since content is not structured, it tends to be resistant to automated analysis. It also includes slang, synonyms, sarcasm, jargon, irony, ambiguity, anaphora, multiple meanings for the same word, and a host of vagaries not found in structured data. Concept Searching technologies automatically resolve these issues by the generation of multi-term semantic metadata.

The objective was to analyze and extract relevant information to improve geological decision making in the selection or abandonment of drilling sites. Access to the project included 5,000 geophysicists globally and required the integration of 52 taxonomies. The end result of the analysis represented millions of dollars in savings to the company.

When coupled with Concept Searching technologies, content analytics provides the ability to find patterns in information such as semantic models, text analytics, or graphical representations of data that provide new insights from large repositories of information.

The solution included the implementation of conceptClassifier for SharePoint, to provide the underlying semantic and geotagging framework to improve search in a SharePoint environment. Users were able to use the core technologies to extract highly specific and relevant information from a large corpus of documents linking 52 taxonomies, for analysis and improved decision making.

Before content analysis, the products are used to narrow the data set to include highly granular information, and remove irrelevant content and noise. Examples of content include geospatial data, documents, emails, scanned paper, databases, images, and videos. Before analysis, Concept Searching recommends performing content optimization, which is a process that eliminates content that is no longer needed or contains no value. Once the data set is clean, analysis is far more accurate and usable in aggregating the information needed to make business decisions.

Best practices include the refining of the data set and implementing a controlled data ingestion process. Policies that reinforce the behaviors necessary for effective information management should be instituted, then content analytics used to create the business case to solve real business problems and encourage organizational support.

Information that is related and inter-related will be identified, generating a far richer base from which analysis can occur. Subject-matter experts can easily refine the content set for analysis. Using standard reporting tools, or any business intelligence tool, information can be generated for analysis. Concept Searching technologies and solutions are easy to use, rapidly deployed, and are designed for use by business professionals rather than IT teams.

The company was able to save millions of dollars through the identification or elimination of drilling sites. It continued to solve business problems using Concept Searching technologies, aggregating information from a large number of repositories to provide an integrated search capability across claims data and financial spend data, delivering consolidated geographic-based reports. The technologies were used to intersect operations and environmental data with claims data, to confirm the economic impact caused by oil spills and to check the validity of claims. It also used the technologies to analyze an array of environmental, operational, and remote sensing data, which scientists needed to access in order to understand the ecosystem, environmental impact, and remediation.

  • Easily used by subject-matter experts, intuitive, and requires little training
  • Ability to extract and aggregate content from multiple taxonomies and repositories
  • Finds unknown content, making it available for analysis
  • Identifies inter-related and intra-related content automatically
  • Provides businesses gains, through greater visibility into asset and resource allocation
  • Creates a clean data set, by eliminating noise and content of no value
  • Integrates with all data and provides access to unlimited content sources
  • Includes a multitude of file types, including 2D and 3D images, drawings, and structured data
  • Platform agnostic
  • Eliminates end user tagging
  • Improves insight and encourages innovation in data purpose and patterns
  • Can be used by any business or artificial intelligence applications for manipulation and analysis

Ask a Question

Leave your details and one of our consultants will get back to you.

Search Webinar