Go to the NASA Homepage
Search >
Click to Search
Human Systems Integration Division homepageHuman Systems Integration Division homepage Organization pageOrganization page Technical Areas pageTechnical Areas page Outreach and Publications pageOutreach and Publications page Contact pageContact page
Human Systems Integration Division Homepage
Outreach & Publications Sidebar Header
Go to the Outreach & Publications pageGo to the Outreach & Publications page
Go to Awards pageGo to Awards page
Go to News pageGo to News page
Go to Factsheets pageGo to Factsheets page
Go to Multimedia pageGo to Multimedia page
Go to Human Factors 101 pageGo to Human Factors 101 page
What is Human System Integration? Website
Publication Header
New contextual data mining technologies  (2000)
Abstract Header
selected contexts, and ranks the narratives on their relevance to the keywords in context. It displays those narratives with their relevant sections highlighted, and also displays the criteria used to determine relevance. QUORUM/Perilog phrase search retrieves narratives that contain one or more user-specified phrases, even hundreds of phrases, and ranks the narratives on their relevance to the phrases. It displays the narratives with the phrases highlighted, and also shows near-matches to the query phrases. QUORUM/Perilog phrase generation creates a list of phrases, from the database of text, that contain a user-specified word or phrase. QUORUM/Perilog phrase discovery finds phrases that are related to topics of interest. For example, a query on the topic of "fatigue" produces results including: "rest period", "reduced rest", "crew rest", "continuous duty", "crew scheduling", "duty period", "rest periods", "reserve or standby", and many others. Phrase discovery is useful for gaining a better understanding of the topics contained in a database. In addition, phrase generation and phrase discovery are particularly useful for finding query phrases for input to QUORUM/Perilog phrase search.

The new data mining methods have been successfully tested on the tens of thousands of narrative incident reports in the database of the Aviation Safety Reporting System (ASRS). The ASRS is a national clearinghouse of aviation safety incident reports supported by the FAA and managed by NASA. The new technologies completed this year will support system-wide ASRS analyses for government, industry, and academic researchers.

The new data mining methods are applicable to a wide variety of text, and several analyses have been done as demonstrations. For example, one analysis involved reports of safety-related incidents that occurred during ground maintenance of the Space Shuttle. In addition, a sample of incidents from the electric power generation industry was analyzed.

Commercialization of the four new data mining methods began with the submission of four formal invention disclosures. The Ames Commercial Technology Office (CTO) reviewed the disclosures, and also hired contractors to conduct a search for prior art and an analysis of the commercial potential of the new technologies. Encouraged by the results, the CTO established a contract with an outside law firm to write four patent applications. The four patent applications are based on nine "inventive concepts", each of which could be an independent patent if the intent were to maximize the number of patents, as is often done in start-up companies. The writing of the patent applications was still in progress at the end of Fiscal Year 2000. After approval by Ames and Headquarters, completed applications are submitted to the United States Patent Office. Meanwhile, the CTO is developing a commercial licensing strategy for the new technologies.

The new methods have been documented in the paper, "Searching the ASRS database usingQUORUM keyword search, phrase search, phrase generation, and phrase discovery". The paper has been approved for publication pending submission of the patent applications.
Private Investigators Header
Authors Header
Groups Header
Keywords Header
contextual data mini, database, phrase search
References Header
Download Header
Adobe PDF Icon  McGreevy_RT2000.pdf (Download Acrobat Reader Click to download Adobe Acrabat Reader)
  (45KB) (application/pdf)
Go to the First Gov Homepage
Go to the NASA - National Aeronautics and Space Administration Homepage
Curator: Phil So
NASA Official: Jessica Nowinski
Last Updated: August 15, 2019