Expert systems and the CPI product substitution review: A needs analysis for the US Bureau of Labor Statistics
This report presents recommendations relative to the use of expert systems and machine learning techniques by the Bureau of Labor Statistics (BLS) to substantially automate product substitution decisions associated with the Consumer Price Index (CPI). Thirteen commercially available, PC-based expert system shells have received in-depth evaluations. Various machine learning techniques were also reviewed. Two recommendations are given: (1) BLS should use the expert system shell LEVEL5 OBJECT and establish a software development methodology for expert systems; and (2) BLS should undertake a small study to evaluate the potential of machine learning techniques to create and maintain the approximately 350 ELI-specific knowledge bases to be used in CPI product substitution review.
- Research Organization:
- Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
- Sponsoring Organization:
- USDOE, Washington, DC (United States)
- DOE Contract Number:
- AC05-84OR21400
- OSTI ID:
- 10124152
- Report Number(s):
- ORNL/TM-11987; ON: DE92008237
- Resource Relation:
- Other Information: PBD: Feb 1992
- Country of Publication:
- United States
- Language:
- English
Similar Records
Report Briefs: Publications of the Energy Division, Oak Ridge National Laboratory, 1999
Explainable Artificial Intelligence Technology for Predictive Maintenance