Case Retrieval for Product Configuration Based on NRS and KNN

Abstract:

Article Preview

The accuracy and efficiency of case retrieval decline with individuality and diversity of customer needs in current product configuration design. To cope with this problem, this paper proposes the method of case retrieval based on NRS and KNN. Firstly, setting neighbor value can effectively avoid errors generated by discretization of continuous attributes in classical RS. Secondly, NRS is used for redundant reduction and weight allocation of attributes, and to calculate distance of each sample points by weighted distance. Thirdly, K similar cases are selected according to Euclidean distance with weight, then the case that meet customer needs best is acquired. Lastly, product configuration of coal mining machine is used to illustrate that accuracy and efficiency of the method are superior to the existing methods of case retrieval.

Info:

Periodical:

Advanced Materials Research (Volumes 139-141)

Edited by:

Liangchi Zhang, Chunliang Zhang and Tielin Shi

Pages:

1525-1529

DOI:

10.4028/www.scientific.net/AMR.139-141.1525

Citation:

T. Xi et al., "Case Retrieval for Product Configuration Based on NRS and KNN", Advanced Materials Research, Vols. 139-141, pp. 1525-1529, 2010

Online since:

October 2010

Export:

Price:

$35.00

In order to see related information, you need to Login.

In order to see related information, you need to Login.