Title :
COPRA: Computer Operations Problem Resolution Assistant
Author :
Shutt, T. Scott ; Simoudis, Evangelos
Author_Institution :
Internal Revenue Service, Washington, DC, USA
Abstract :
The authors are developing a case-based reasoning (CBR) system called COPRA to help computer operations personnel at Internal Revenue Service (IRS) service centers to recover from problems that occur during the processing of databases which contain taxpayer information. The case base in this system contains prototypical cases developed from experiences of senior personnel from the IRS National Office in Washington, DC, and the IRS service center in Fresno, CA. The performance of the prototype system has been evaluated in a laboratory setting with regard to noise and the size of its case base. This evaluation showed that, for the particular class of problems that the current COPRA prototype is expected to address, both the size of the case base and the amount of noise in it had a significant effect on retrieval. The system was implemented using the CBR Express shell
Keywords :
case-based reasoning; finance; help systems; public administration; tax preparation; utility programs; CBR Express shell; COPRA; Computer Operations Problem Resolution Assistant; IRS National Office; Internal Revenue Service; case-based reasoning; computer operations personnel; databases; prototypical cases; service centers; taxpayer information; Artificial intelligence; Data analysis; Hardware; Knowledge acquisition; Laboratories; Performance analysis; Personnel; Problem-solving; Prototypes; Trademarks;
Conference_Titel :
Artificial Intelligence for Applications, 1993. Proceedings., Ninth Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-8186-3840-0
DOI :
10.1109/CAIA.1993.366654