Integrating Natural Language Processing and Information Retrieval in a Troubleshooting Help Desk

  • Authors:
  • Peter G. Anick

  • Affiliations:
  • -

  • Venue:
  • IEEE Expert: Intelligent Systems and Their Applications
  • Year:
  • 1993

Quantified Score

Hi-index 0.00

Visualization

Abstract

Digital Equipment Corporation provides telephone and electronic support to its customers through a world wide network of support centers and field offices. In addition to troubleshooting customer problems, Digital's service and support specialists share their knowledge and experiences with other specialists by writing up their problems and solutions and making them available in a worldwide online information base. By sharing this knowledge among a large body of physically distributed support engineers, customer problems that have been previously encountered can be resolved much more quickly, thereby increasing customer satisfaction and decreasing Digital's cost of providing service. Over the past decade, the range of products supported by Digital's specialists has grown steadily and now includes non-Digital products as well. This growth has been matched by a massive increase in the size of the on-line information base, which has surged from less than 10,000 articles into the hundreds of thousands. As the size and importance of the database has grown, the customer support centers have improved their software's data management and information retrieval capabilities. The author argues that natural language processing plays an important role.