Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/3998
Title: | Evaluation of ordinal attributes at value level | Authors: | Robnik-Sikonja, Marko VANHOOF, Koen |
Issue Date: | 2007 | Publisher: | SPRINGER | Source: | DATA MINING AND KNOWLEDGE DISCOVERY, 14(2). p. 225-243 | Abstract: | We propose a novel context sensitive algorithm for evaluation of ordinal attributes which exploits the information hidden in ordering of attributes' and class' values and provides a separate score for each value of the attribute. Similar to feature selection algorithm ReliefF, the proposed algorithm exploits the contextual information via selection of nearest instances. The ordEval algorithm outputs probabilistic factors corresponding to the effect an increase/decrease of attribute's value has on the class value. While the ordEval algorithm is general and can be used for analysis of any survey with graded answers, we show its utility on an important marketing problem of customer (dis) satisfaction. We develop a visualization technique and show how we can use it to detect and confirm several findings from marketing theory. | Notes: | Univ Ljubljana, Fac Comp & Informat Sci, Ljubljana 1001, Slovenia. Univ Hasselt, Dept Econ, B-3590 Diepenbeek, Belgium.Robnik-Sikonja, M, Univ Ljubljana, Fac Comp & Informat Sci, Trzaska 25, Ljubljana 1001, Slovenia.Marko.Robnik@fri.uni-lj.si koen.vanhoof@uhasselt.be | Keywords: | attribute evaluation; ordinal attributes; attribute values; visualization; marketing | Document URI: | http://hdl.handle.net/1942/3998 | ISSN: | 1384-5810 | e-ISSN: | 1573-756X | DOI: | 10.1007/s10618-006-0048-4 | ISI #: | 000245281600002 | Category: | A1 | Type: | Journal Contribution | Validations: | ecoom 2008 |
Appears in Collections: | Research publications |
Show full item record
SCOPUSTM
Citations
11
checked on Sep 2, 2020
WEB OF SCIENCETM
Citations
9
checked on Oct 12, 2024
Page view(s)
52
checked on Jul 31, 2023
Google ScholarTM
Check
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.